Subido por Yudivián Almeida Cruz

BookofabstractIWOR2015Cuba.

Anuncio
11th INTERNATIONAL WORKSHOP ON
OPERATIONS RESEARCH
Havana; March 10-13, 2015
Main Themes:
OR & Human Welfare: Health Environment and
Education.
Organized by:
Universidad de La Habana.
Université Paris 1, Panthéon-Sorbonne.
Cosponsors:
Asociación Latinoamericana de Investigación
Operativa.
Sociedad Cubana de Matemática y Computación (SC
Investigación Operacional).
Oficina del Historiador de la Ciudad.
American Statistical Association.
.
Committees
Organizing Committee
Raúl Guinovart Diaz: Decano, Facultad de Matemática y Computación, Genry Pérez, Saraí
Moreno.
Program Committee
S. Allende [Chair, Havana], L. Alvarez [Havana], A. Arcos [Granada], M.L. Baguer [Havana],
C. Bouza [Havana], G. Bouza [Havana], R. Caballero [Málaga], B. Cornet [Paris 1-Kansas],
M. Cortés [Cienfuegos], D. Covarrubias [Chilpancingo], M. Cottrell [Chair, Paris 1], J.
Cochran [Alabama] J. Daduna [Berlin], Espín [Havana], P. Gaubert [Paris 12], J. Guddat
[Honor Chair, Berlin], C. Hardouin, [Paris Ouest Nanterre], S. Hernadez [Xalapa], J. C. Juvier
[Las Palmas], A.M. Lara [Granada], M. A. León [Pinar del Río], A. Marrero [Havana], A.
Mesejo [Havana], M. Negreiros [Forteza], M. Nicado [Havana], M. Olteanu [Paris 1], J.
Otero [Havana], J. Rückman [Norway], A. Ruiz [Havana], Ch.Tammer [Halle], L. Sandoval
[Puebla], V. Sistachs [Havana], J.L. Verdegay [Granada], B-A. Wickström [Berlin], A. Xavier
[Rio de Janeiro].
Index
Forewords…………………………………………………………4
Lecturers…………………………………………………………...5
Timetable...………………………………………………………....6
Plenary Lectures……………………………………………............12
Main Lectures……..…………………………………………….....15
Contributions……………………………………………………....18
Plenary Speakers………..………...…….………………………….42
Main Speakers………………….………………………………….44
List of Participants………………………………………………....45
Lecturers
Plenary speakers
M. Ahlheim (Germany).
J. Cochran (USA).
M. Negreiros (Brazil).
J. Rynkiewicz(France).
Main speakers
L. R. Keller (USA).
M. Oliveira (Brazil).
J. Rynkiewicz.
Timetable
Tuesday Morning
8:30-9:45: Registration, Salón del Monte, Ambos Mundos Hotel.
9:45-10:30: Opening Address, Salón del Monte, Ambos Mundos Hotel.
11:00-13:00: Gathering Party, Restaurant La Dominica.
Tuesday Afternoon
(to be held at Colegio San Gerónimo de la Habana)
13:00-14:00: M. Ahlheim: The economic valuation of changes in land use towards more
sustainability – An empirical example from southwest China. (Opening Lecture). Room 1.
Introducer: B. A. Wickström.
14:00-14:20: Coffee Break.
Time
Room 1
Room 2
14:20-15:40
Mathematical Models I
Applications to Economy II
Session: Mathematical Models I. Room 1.
Chair: A. Leon.
Time
Speaker
Talk
14:20-14:40
Koné, M.I.
Pontryagin Maximum Principle in Optimal Control with delay in
state space.
14:40-15:00
Rabenoro, T. Leveraging experts’ knowledge to improve diagnostics on aircraft
engine.
15:00-15:20
Skala, V.
A practical use of radial basis functions interpolation and
approximation.
15:20-15:40
Bouza, G.
A comparison between smoothing and regularization approaches
for MPCC.
Session: Applications to Economy II. Room 2.
Chair: A. F. Arguello Ortiz.
Time
Speaker
Talk
14:20-14:40
Salgado, L.
Evaluación de la pobreza, desarrollo comunitario o
desarrollo local y el municipio.
14:40-15:00
Caamal Cauich. I.
Seasonality of Mexico Persian lime prices.
15:00-15:20
Garcia Rodríguez, J.
Pobreza y otros determinantes socio-económicos y
F.
financieros del acceso a vivienda de interés prioritario en
América Latina. El caso colombiano.
Caamal
Cauich.
I.
Behavior of agricultural interest rates in Mexico.
15:20-15:40
15:40-16:00 Priego Hernández, O. Los escenarios de la globalización y la pobreza en México.
Wednesday Morning
9:00-10:00: J. Cochran: The education generation gap: confronting and embracing the challenges
that growing computing power and new technology present to Applied Mathematics Education.
(Plenary Lecture). Room 1.
Introducer: Sira Allende.
10:00-10:20: Coffee Break.
Time
Room 1
Room 2
14:20-15:40
Economics and Decisions Models
Teaching Effectiveness Colloquium I
Session Economics and Decisions Models. Room 1.
Chair: M. Ahlheim.
Hour
Speaker
Talk
10:20-10:40
Collier, I. L.
Distribution and redistribution in a downton abbey
economy.
10:40-11:00
Gonçalo B. P. R.
Calculate characterization of monopsonic degree into the
waste solid market in large metropolitan of Brazil.
11:00-11:20
Wickström, B. A
Language rights: a welfare-economics approach.
11:20-11:40
Klunover, D.
The theory of sequential lotteries.
11:40-12:00
Sarin, R. K.
Guided decision processes.
Teaching Effectiveness Colloquium I, Room 2.
Chair: J. Simon.
Hour
Speaker
Talk
10:20-11:00
Cochran, J. J Active learning: reengaging and reenergizing your students and
their instructor. (Main Lecture)
11:00-11:20
Baltazar, J.C. Employment of new graduates in the Mexican labor market
(case study, University of the State of Mexico).
11:20-11:40
Gaviño, G.
Econometric & statistical response model applied to socioeconomical careers at UAEM. C. U. Valle de México.
11:40-12:00
Vázquez, Y.
Hiperentorno basado en objetos de aprendizaje destinado a la
asignatura análisis de datos (Estadística) para la carrera de
Cultura Física.
12:00-12:20
Allende, S.
Teaching OR in different contexts
12:00-14:00: Lunch Break.
Wednesday Afternoon
Time
14:00-16:00
Room 1
Análisis de Grandes Volúmenes de
Datos Textuales
Room 2
Teaching Effectiveness Colloquium II
Conversatorio Internacional: Análisis de grandes volúmenes de datos textuales. Room 1.
Coordinador: S. Hernández.
Panelistas: A. Agüero, M. Laborde y M. Oller.
Teaching Effectiveness Colloquium II, Room 2.
Chair: G. Gaviño.
Hour
Speaker
Talk
14:00-14:40
Keller, L.R.
Teaching multi-objective multi- stakeholder decision modeling
with cases (Main lecture)
14:40-15:00
Reyes, H.
Modelación estadística sobre el egreso de universitarios de
ciencias exactas de la FCFM-BUAP.
15:00-16:00: Poster Session 1, Room 2.
Chair: Y. Almeida.
Speaker
Talk
Estévez, S.
Influencia del uso de medidas de similaridad semántica para el proceso de
minería de opinión en Twitter.
Las transformadas de Watershed y Waterfall para la segmentación de
Baguer, M.L.
imágenes de la aduana.
Estévez, S.
Wikipedia as a corpus for Text Mining.
Thursday Morning
9:00-10:00: Rynkiewicz, J Assessment of the influence of education level on voting intention for
the extreme right in France (Plenary Lecture). Room 1.
Chair: C. Bouza.
10:00-10:20: Coffee Break.
Time
10:20-12:30
Room 1
Human Health Studies I
Room 2
Computer Vision and Graphics
Session Human Health Studies I. Room 1
Chair M. Oliveira
Hour
Speaker
Talk
10:20-11:00
Sombra, F.
Processing: applications in health care and remote sensing of
environment.
11:00-11:20
Uranga, R.
Simulación para evaluar datos faltantes en modelos de
transición en el contexto de ensayos clínicos.
11:20-12:30
Panel
Conversatorio: Los métodos matemáticos en la Medicina.
Session Computer Vision and Graphics. Room II
Chair A. Mesejo.
Hour
Speaker
Talk
10:20-10:40
León, A.
Movement detection using Wavelet.
10:40-11:00
Mesejo, D.
Contrast enhacement for digital mammography images
based on wavelets.
11:00-11:20
Piad, A.
E-buffer. A GPU friendly space partitioning based on
bounding volume hierarchy of empty-spaces suitable for
raycasting applications.
11:20-11:40
Kauffmann, C.
A GPU frameworks to compute supervoxels in large medical
images.
12:00-14:00: Lunch Break.
Thursday Afternoon
Time
14:00-12:30
Room 1
Human Health Studies II
Room 2
Poster Session
Session Human Health Studies II, Room 1.
Chair F. Sombra.
Hour
Speaker
Talk
14:00-14:40
Oliveira, M.
Evolving an intelligent framework for decision- making
process in E-health systems. (Main Lecture).
14:40-15:00
Lage, A.
Inference of source origin in epidemics and water distribution
systems.
15:00-15:20
López, L.
Prognostic factors for cervical cancer patients in Tabasco,
México.
15:20-15:40
Betancourt, J. Entrenamiento en análisis epidemiológicos para estudiantes
de Medicina.
14:00-15:00: Poster Session 2, Room 2.
Chair: V. Sistachs.
Arcos, M del M.
Rueda, M del M.
Tuero A.
Viada, C.
Gender analysis of medical professionals in Spanish public
universities.
Un estudio estadístico sobre el uso problemático del móvil y
personalidad.
Meta-analysis of safety and efficacy in patients with diabetic foot
ulcers. International projection of Heberprot-P®.
Tiempo al deterioro de los scores de calidad de vida como una
modalidad de análisis longitudinal para pacientes con cáncer de
cabeza y cuello tratados con Nimotuzumab.
15:00-16:00: Poster Session 3, Room 2.
Chair: V. Sistachs.
Arcos, A.
León López, E.
Sistachs, V.
Eduardo, A.
The use of landline and cell-phone in health surveys.
Estudio del pronóstico de la gravedad del paciente: neumonías
adquiridas en la comunidad. Hospital General Docente Iván
Portuondo, San Antonio de los Baños
Promedio Bayesiano de Modelos en Salud: una revisión.
Análisis de desempeño académico de los estudiantes en la
asignatura de estadística en la Universidad Pedagógica de Benguela
– República de Angola.
Friday Morning
9:00-10:00: Negreiros, M. Capacitated centered clustering problem: applications, models, exact
and metaheuristic methods (Plenary Lecture). Room 1.
Chair: C. Bouza.
10:00-10:20: Coffee Break.
Time
Room 1
Room 2
14:00-12:30
Applications of OR
Statistical models
Session Applications of OR. Room 1
Chair: A. Ruiz.
Hour
Speaker
Talk
10:20-10:40
Agüero, L.
Aplicación del método CDS secuencial para el rediseño de las
rutas de distribución en una fábrica productora de helado.
1020-1040
Tamayo, A.
Diseño de un procedimiento para la planificación y control de la
producción haciendo uso de herramientas matemáticas.
10:40-11:00
García, D
Herramienta computacional para la planificación de rutas en el
problema de enrutamiento de vehículos considerando una flota
heterogénea.
11:00-11:20
Limia, L.
Una estrategia por vecindades para el problema de
enrutamiento de vehículos con compartimentos.
11:20-11:40
Marrero, A.
Algoritmo genético semidifuso. Aplicación a un modelo de
detección del VIH en Cuba.
11:40-12:00
Estévez, S.
Influence in Twitter: a proposal based on clustering.
12:00-12:20
Nguyen, B.
A dual method to solve a quadratic program derived from a
distance metric learning problem.
Session Statistical models. Room 2.
Chair. V. Sistachs.
Hour
Speaker
10:00-10:20
Matías Castillo, B.C.
10:20-10:40
Monterrey , P.
10:40-11:00
Castro Alva, J.J.
11:00-11:20
11:20-11:40
Zabala Corrales, J. J.
Echevarria Molina, C.
11:40-12:00
Bouza, C.
Talk
Análisis de correlación canónica usando algoritmos
genéticos.
Caracterización de la distribución de los valores p de
las pruebas paramétricas en poblaciones normales.
Cálculos de los tamaños de prueba y niveles de
significancia ajustados para la prueba de Blackwelder.
Population sampling fuzzy.
Statistical analysis in the evaluation of intensity of
knowledge work.
Ratio type estimation using the knowledge of the
auxiliary variable for ranking and estimating.
Friday Afternoon
14:00-16:00: Meeting of the network RIDECA:
Coordinator C. Bouza.
Activities:
Presentation of the books:
- Pobreza y desigualdad. Una mirada desde la economía y la investigación operacional,
J. M. Sautto.
-
Experiencias en la modelación de la toma de decisiones en la salud humana, medio ambiente y
desarrollo humano, tomo 1. J. F. Garcia.
Plan of the activities of 2015-2016, C. N. Bouza.
Plenary Lectures
THE ECONOMIC VALUATION OF CHANGES IN LANDUSE
TOWARDS MORE SUSTAINABILITY – AN EMPIRICAL EXAMPLE
FROM SOUTHWEST CHINA
Michael Ahlheim
Volkswirtschaftliches Institut , Universität Hohenheim, Germany.
Especially in developing and emerging countries we often observe a fundamental conflict
between the economic exploitation of land and environmental preservation. Projects aiming
at environmental preservation have to be organized by government institutions who also
have to pay the project costs out of public funds which could also be used for alterative
purposes. In this context the question arises if a certain preservation project is worth its
costs from a social point of view or if the public money should better be used otherwise.
While the project costs can be calculated on the basis of market prices in most cases this is
not possible for the social benefits accruing from such a project. Therefore, special
valuation techniques must be applied for their assessment. In this presentation one of these
techniques, the so-called Contingent Valuation Method (CVM), will shortly be introduced
and then an empirical example from Xishuangbanna in Southwest China will be presented.
The cultivation of rubber trees in Xishuangbanna Prefecture in China’s Yunnan Province
has triggered an unprecedented economic development but it is also associated with severe
environmental problems. Rubber plantations are encroaching the indigenous rainforests at a
large scale and a high speed in Xishuangbanna. Many rare plant and animal species are
endangered by this development, the natural water management is disturbed and even the
microclimate in this region has changed over the past years. The present study aims at an
assessment of the environmental benefits accruing from a reforestation project partly
reversing the deforestation that has taken place over the past years. To this end a
Contingent Valuation survey has been conducted in Xishuangbanna to elicit local residents’
willingness to pay for this reforestation program that converts existing rubber plantations
back into forest. The sum of the willingness to pay of all people affected by this program
will be interpreted as its social value.
It is shown that local people's awareness of the environmental problems caused by
increasing rubber plantation is quite high and that in spite of the economic advantages of
rubber plantation there is a positive willingness among the local population to contribute
financially to a reduction of existing rubber plantations for the sake of a partial restoration
of the local rainforest. These results could be used for the practical implementation of a
PES (Payments for Eco‐System Services) system for reforestation in Xishuangbanna.
THE EDUCATION GENERATION GAP: CONFRONTING AND
EMBRACING THE CHALLENGES THAT GROWING COMPUTING
POWER AND NEW TECHNOLOGY PRESENT TO APPLIED
MATHEMATICS EDUCATION
James J. Cochran
University of Alabama, USA.
It is almost impossible to overstate the continual (perhaps even continuous) impact and
potential impact of increasingly powerful computers and new technology have on
education. During our childhoods we developed methods for processing information and
understanding concepts that were foreign to our instructors, and now we are faced with the
same dilemmas we presented to our instructors; our students have developed methods for
processing information and understanding concepts that are foreign to us. It is important for
us to understand that this general challenge is not new, and only the circumstances and
surroundings change from generation to generation.
This can be frustrating for our students and maddening for us, and it leads to many
important questions. How do we at the very least cope with, or more hopefully embrace and
efficaciously utilize, these new technologies and alternative methods for processing
information and understanding concepts? Are these alternative methods for processing
information and understanding concepts really new? Would we have developed similar
methods for processing information and understanding concepts if the same technology had
been readily available during our formative years?
In this talk Professor Cochran will address issues that arise when integrating new
technology into the classroom and possible strategies for using new technology to make
applied mathematics instruction more effective. He will demonstrate several simple uses of
technology for educational purposes and will discuss several of his experiences, both
positive and negative, with integrating technology into his instructional efforts.
CAPACITATED CENTRED CLUSTERING PROBLEM:
APPLICATIONS, MODELS, EXACT AND METAHEURISTIC
METODS.
Marcos Negreiros*, Pablo Batista*, Albert Muritiba** & Augusto Palhano*
*Universidade Estadual do Ceará.
**Universidade Federal do Ceará.
This talk considers the Capacitated Centered Clustering Problem, a recent proposed NPhard problem that can be applied to sales force geographical distribution, division of areas
for garbage collection, wholesale distribution, sanitary agent coverages, routing newspaper
delivery for subscribers, agro-food supply chain network, allocation of emergency shelters
to protect local people during possible natural disasters, computational biology and facility
location, etc. We show mathematical models and their effectivity to be solved by state-ofthe-art solvers, new formulations and applications for heterogeneous clustering in TI and
Environment. Finally we show exact and most recent strategies used by metaheuristic
methods designed which obtain the best results for the majority of difficult instances of the
problem proposed by the literature.
ASSESSMENT OF THE INFLUENCE OF EDUCATION LEVEL ON
VOTING INTENTION FOR THE EXTREME RIGHT IN FRANCE.
Joseph Rynkiewicz*, Mohamed Raouf Benmakrelouf** & Wafa Karouche
*Université de Paris I, Paris – France.
**2-USTHB, Algiers - Algeria
In France, the Front National has been a growing political party in the last 30 years. After
years of stagnant economy, French voters have come to mistrust the political elite or any
decisions from the European Union, and have been increasingly receptive the Front
National straight-talking approach. The most consistent findings in social research on
ethnic attitudes is the negative association between educational attainment and ethnic
prejudice: People with higher education are less prejudiced toward ethnic out groups than
those with lower education. We might expect, then, that high education will generally
prevent people from voting for the extreme right, regardless of their position in the labor
market. This point of view, however, is not unanimous and there is an intellectual elite
within the extreme right, especially among angry academic white males who are uneasy
with the gains of feminism and believe in a left-wing media conspiracy. Hence the link
between Extreme Right voting and education is not so obvious. The aim of this paper is
first to build a model which takes into account the causality structure of variables (Bayesian
Network) and secondly to assess the influence of education on the voting intention for
extreme right-wing party by taking into consideration all the possible confounding
variables
Main Lectures
ACTIVE LEARNING: REENGAGING AND REENERGIZING YOUR
STUDENTS AND THEIR INSTRUCTOR!
James J. Cochran
University of Alabama, USA.
We in the operations research community understand that OR and Statistics are inherently
interesting, relevant, important, and enjoyable disciplines - unfortunately many of our
students (particularly those in introductory OR and Statistics courses) don't seem to share
this understanding with us! So how do OR and Statistics instructors help students
appreciate that OR and Statistics are interesting and relevant and important and enjoyable?
Professor Cochran discusses several classroom cases and active learning exercises he has
developed and regularly uses to accomplish this goal when teaching introductory OR and
Statistics courses.
Throughout this session Professor Cochran will emphasize his points with live
demonstrations and discussions of several interesting and novel active learning exercises
and cases. Card tricks, classroom versions of television game shows, and a teaching case
with integrated active learning will be featured. Because many of these exercises are easily
transferable across topics, instructor/classroom styles, cultures, national borders,
institutions, faculties, programs, and class sizes, it is very likely you will walk away from
this session with ideas on how to improve your own teaching (indeed, Professor Cochran
will be very disappointed if you don't!). Be prepared – audience participation is an integral
part of this session, and Professor Cochran may call on you during the session!
EVOLVING AN INTELLIGENT FRAMEWORK FOR DECISIONMAKING PROCESS IN E-HEALTH SYSTEMS
Leonardo M. Gardini*, Carina Oliveira**, Reinaldo Braga**, Ronaldo Ramos**, Luiz O.
M. Andrade*** & Mauro Oliveira**
*State University of Ceará (UECE) - Fortaleza, Brazil.
**Federal Institute of Ceará (IFCE) - Fortaleza, Brazil.
***Federal University of Ceará (UFC) - Fortaleza, Brazil.
This paper presents improvements of LARIISA, a framework that makes use of contextaware information to support decision-making and governance in the public health area.
More specifically, two relevant e-health applications are presented to illustrate the
LARIISA system. The first one uses Bayesian networks in dengue scenarios. The second
application uses ontology to manage home care scenarios. In both cases, the contributions
related to the LARIISA framework include patient health diagnosis provided remotely,
support for decision-making health systems, and context information for context-aware
health systems.
TEACHING MULTI-OBJECTIVE MULTI-STAKEHOLDER
DECISION MODELING WITH CASES
L. Robin Keller* & Jay Simon**
*University of California, Irvine, USA
**Defense Resources Management Institute, USA
Many operations researchers know about the use of decision analysis to decide among
alternative investments (such as pharmacological research and development) using decision
trees with chance nodes to compute expected monetary value of different alternatives. Such
an analysis aims to maximize a single evaluation measure for a single decision maker. We
demonstrate less widely known decision analysis techniques using spreadsheet models of
the multiple objective perspectives of the decision stakeholders. We show how to teach
students to analyze real-life decision problems using case examples and discuss specific
skills students are expected to learn, such as dynamic sensitivity analysis using sliders in
Excel on objectives’ weights, and typical student questions and errors during case
discussion. This methodology has been taught successfully in business courses for both
MBAs (including health care executive MBAs) and undergraduate students.
Sometimes, one objectives hierarchy is suitable for a set of stakeholders, and differences in
opinions across stakeholders can be characterized by differences in the multiple objectives’
weights. Examples include the analysis for the merger of the Operations Research Society
of America and The Institute of Management Sciences to become INFORMS and planning
for protection against radioactive iodine releases in nuclear incidents. In other cases, an
objectives hierarchy will be constructed for each stakeholder because their objectives are so
different that construction of separate hierarchies better represents their divergent
perspectives. Examples include a prostate cancer treatment decision (of former Intel CEO
Andy Grove, his family, his company, and his doctors), a tuna fish supplier source selection
decision (from the perspectives of the StarKist company, environmentalists, and the San
Diego tuna fishing fleet), and the potential siting of a new Home Depot building supply
store.
Sources:
http://pubsonline.informs.org/doi/pdf/10.1287/educ.1090.0066,
http://dx.doi.org/10.1287/ited.1080.0012.
PROCESSING: APPLICATIONS IN HEALTH CARE AND REMOTE
SENSING OF ENVIRONMENT
Fátima Sombra Medeiros
UFC-DETI, Brazil
This talk approaches the use of image processing tools in medical applications and in
synthetic aperture radar (SAR) systems. We have developed methods that assist health
professionals in human posture evaluation and retinal disease detection and identification.
The posture evaluation algorithm is capable of identifying posture problems and assesses
the physiotherapeutic treatment evolution based on a similarity measure. In fact, it is a
suitable and unexpensive solution to help health professional to provide an accurate
diagnosis of postural problems and furthermore it reduces errors due to subjective
evaluation.
Regarding retinal diseases, we have studied the diabetic retinopathy (DR) which is one of
the major complications of diabetes mellitus and furthermore it causes severe damage to the
retina and consequently to the vision. DR may lead to blindness and therefore it is
important to prevent it or early detect and treat it. The diagnosis of DR is performed by
visual analysis of retinal images being exudates (fat deposits) the main patterns traced by a
specialist doctor. It is noteworthy that early diagnosis through regular monitoring when
coupled with proper treatment results in numerous benefits in the prevention of visual
impairment. We will present image processing tools which we have developed for optical
disk, retina vessel network, exudate and microaneurysm detection in retinal images.
Regarding remote sensing of environment applications, we have focused on environmental
disasters by using synthetic aperture radar (SAR) systems, which have played an important
role in remote sensing. These systems provide oil spills detection and monitoring, that
seriously affect the marine ecosystem, providing a more rigorous and effective environment
monitoring. Furthermore, SAR images have considerably contributed to understand
atmospheric phenomena, land use mapping and monitoring, deforestation assessment,
geographic evolution, urban growing rates assessment, agricultural crops monitoring and so
on. The potential damage for the environment and economy of the area at stake requires
that agencies be prepared to rapidly detect, monitor, and clean up any large spill. Remote
sensing of dark spots in the sea is a complex process, due to the simultaneous movement of
radar and spots. The presence of an oil film on the sea surface damps out the small waves
and reduces the rough surface due to the increased viscosity of the top layer and drastically
reduces the measured backscattering energy, resulting in darker areas in SAR imagery. The
interest in appraising texture features in this work becomes from the different rough
degrees presented in SAR images. Oil spill images are characterized by being less rough
when compared to the similar slicks. The diffusion of the electromagnetic waves in the
surface of the sea depends, mainly, on the rough surface which is influenced by the
presence of winds, currents, waves and parameters of the radar, such as incidence angle,
frequency, polarization and resolution. The sea behaves as a specular surface when there
are not waves and winds. However, dark areas might not be oil slicks but merely local wind
effects or natural oil films due to low winds.
Contributions
MOVEMENT DETECTION USING WAVELET
Milene Acosta Pita*, Ángela León Mecías**
* Neurociencias, Cuba.
**Universidad de La Habana, Cuba.
Today the intelligent video surveillance systems are imposed on passive surveillance video
systems, since these can detect moving objects in the initial stage and then process tasks
such as object classification, tracking, behavior description object recognition, among
others. This information can be invaluable for safety in general, this includes for instance
police investigation departments for legal evidence, traffic monitoring and ecosystem
monitoring among others. In this research we intend to apply Discrete Wavelet Transform
on motion detection in sequences of images to monitor the movements of patients who are
conducting rehabilitation exercises without the presence of a physiotherapist, or for
supervision of children who are alone in their woom and surveillance of premises. It is
proposed to make the movement detection completely with wavelets, it means preprocessing and motion detection. The idea of the article "Motion detection using Fourier
image reconstruction" (Du- Ming Tsai, Wei - Yao Chiu, Taiwan, 2008) was followed but
using Discrete Wavelet Transform instead of the Fourier Transform to remove the vertical
line pattern (background) and retain only the foreground in the reconstructed image. A
simple segmentation is then made. For validation of the performed strategy an extensive
experimentation is presented.
APLICACIÓN DEL MÉTODO CDS SECUENCIAL PARA EL
REDISEÑO DE LAS RUTAS DE DISTRIBUCIÓN EN UNA FÁBRICA
PRODUCTORA DE HELADO
Liset Agüero Zardón, Idalianys Urquiola Garcia
*Instituto Superior Politécnico “José Antonio Echeverría”, Cujae. Cuba.
Todo acto de la vida humana está inevitablemente ligado a un proceso social de producción
y no es posible entender ninguna producción sin el consumo y al revés. En la actualidad las
empresas conocen que para ser competitivas no es suficiente ofrecer productos de calidad a
un costo óptimo de producción, sino que además es necesario que los clientes puedan
acceder a ellos en cualquier lugar y situación posible, siempre a un precio razonable;
aspecto en el que la planificación de las rutas de distribución juega un papel fundamental.
La aplicación de técnicas cuantitativas permite optimizar el diseño y/o rediseño de las rutas
de distribución. El presente trabajo propone un algoritmo para el rediseño de las rutas de
distribución en una fábrica productora de helado, cuya aplicación permite la disminución de
los indicadores de consumo, costos de distribución y distancia recorrida.
INFLUENCE EN TWITTER: A PROPOSAL BASED ON
CLUSTERING
Yudivian Almeida-Cruz, Suilan Estévez-Velarde &Ariel Hernández-Amador
University of Havana, Cuba.
Twitter is today one of the most important social networks in the Internet. This platform is
useful for analyzing the influence that some users have on the rest of the social network.
This paper presents a new alternative for estimating a user's influence in Twitter. Instead of
trying to devise a new quantitative measure of influence, we propose the use of clustering
algorithms to determine users that belong to specific influence groups. We identify and
analyze a set of variables, both related to a user and to the structure of the social network
itself that characterize his influence.
TEACHING OR IN DIFFERENT CONTEXTS
Sira Ma. Allende Alonso* C: Bouza* Maclovio Sautto**
*Universidad de la Habana, Cuba.
**Universidad de Guerrero, Mexico.
Undergraduate studies of natural and social sciences include courses on statistics methods
Nowadays, the problems tackled by those sciences and the new technologies demand more
tools of OR. An example in that sense is the use of combinatorial optimization methods in
Molecular Biology. Post-graduate Education has to cover such requirements. Of course,
interests on OR differ from one profession to another, consequently, contents and teaching
strategies should be different. In this contribution we propose general strategies and
configurations of the knowledge for introducing OR contents for education in natural and
social science. The strategy is based in the integration of knowledge of different disciplines.
We propose an organization of the contents based on the study of cases for no mathematic
students, in place of the traditional discipline structure, proper for future mathematicians.
GENDER ANALYSIS OF MEDICAL PROFESSIONALS IN SPANISH
PUBLIC UNIVERSITIES
María del Mar Arcos
Universidad de Granada. Spain.
In this paper we present the results of a gender study that has been carried out in order to
know the status of women within the community of researchers and teachers in Spanish
universities in the area of health sciences, and especially in medical degrees. The
methodology used for research has been the census study. The data has been provided by
the National Statistical Institute of Spain (INE). The analysis was performed disaggregating
by professionals from different areas of knowledge related to medicine, with the purpose of
finding out in a more accurate way the situation of women in this group.
In addition, there have been carried out the counting, not only for the total of teaching and
research staff, but also distinguishing each professional according to their category.
Based on the results obtained, one can conclude that women are in an unfavorable position
in relation to teaching and researching at the Spanish public universities in the area of
health sciences.
THE USE OF LANDLINE AND CELL-PHONE IN HEALTH SURVEYS
Maria del Mar Arcos & Antonio Arcos**
Universidad de Granada. Spain.
This work is a review of several estimation methods for dual-frame designs in the particular
context of telephone surveys. Due to the recent increase of the number of people who has
no landline phone but only mobile phone, it is very important nowadays that telephone
surveys incorporate landline together with cell-phone samples. Otherwise, large bias may
appear in the estimations. Given that the two frames made of landline owners and cellphone owners intersect, some adapted estimation methods have to be used and there are
several in the literature. The paper presents the different existing methods and compares
these methods on a real health survey concerning the attitude of the Andalusian population
regarding the public health system. The results suggest that the use of multiple frames
might be useful in other health surveys where good estimates are wanted for both the whole
population and particular subgroups at the same time.
ECONOMETRIC & STATISTICAL RESPONSE MODEL APPLIED
TO SOCIO-ECONOMICAL CAREERS AT UAEM. C. U. VALLE DE
MÉXICO.
Juan Carlos Baltazar Escalona, Gabriela Gaviño Ortiz & Ángel Manuel Sánchez,
Mejorada Zapata
UAEM. Centro Universitario Valle de México, Mexico.
This research is based on an econometric study to evaluate and predict on a qualitative
response model, how the academic performance of students in 9th semester has been
affected by variables (academic and others), during their studies in the undergraduate
programs for socio-economical careers, implemented at the UAEM Centro Universitario
Valle de México. The whole extent of the present study will explain and determinate how
the academic performance has been affected by those factors (13 independent variables
considered), and therefore analyze the impact and level of significance that it means, to
explain by a regression model the findings. Other output to obtain from this study is the
statistical analysis obtained from the data input. The information was obtained by a
questionnaire delivered to the whole population on 9th semester at UAEM CU Valle de
México, during September 2014. Once the information was obtained it was captured on a
SPSS software to obtain the output to do the next steps. The explanatory variables were
classified as: academic, economic and socio-demographic. The purpose is to identify which
of the analyzed variables determine the academic performance of students and establish
some recommendations to UAEM authorities to have a better students’ performance.
The main results of the econometric analysis found that six of the nine academic variables
were statistically significant. In this sense, the analysis corroborates that the materials
disapproving, loading materials, career choice, career expectations and hours of study
influence academic performance. In terms of economic variables, as well as sociodemographic, were not statistically significant, that is, do not affect the academic
performance of a student at UAEM CU Valle de México. What strikes the study is that only
academic variables explain academic performance. Moreover, contrary to expectations,
variables such as age, marital status and employment status not explain that performance.
Other variables whose interest was to find out what effect could lead to academic support,
study option, prestige, because of registration and location were not relevant to the present
study. It can be concluded that the problem of academic performance is due to internal
issues rather than external or extraneous issues of the University. The varying academic
performance focuses more on personal problems of each student of academic nature and
less on the study programs and curricula of the University. The implementation of
educational policies in order to improve academic performance and / or reduce the low
level of qualification should be aimed, among other important aspects, in reviewing the
course load.
CONVERSATORIO INTERNACIONAL SOBRE EL ANÁLISIS DE
GRANDES VOLÚMENES DE DATOS TEXTUALES
Panelistas: Daniel Barredo Ibáñez*, Martín Oller Alonso*, Amaro La Rosa** &
Sergio Hernández**
*Universidad de las Américas, Ecuador.
** Universidad Jaime Bausate y Meza, Perú.
***Universidad Veracruzana, México.
La sociedades contemporáneas registran grandes volúmenes de datos textuales: sea en las
redes sociales; en las plataformas digitales; en los contenidos de los medios de
comunicación; en las transcripciones de las entrevistas; en libros, artículos y foros, por citar
los espacios más representativos, se acumula una vasta cantidad de información. La
aplicación de técnicas como el Análisis Estadístico de Datos Textuales permite abordar de
una forma relativamente sencilla la clasificación e interpretación de todos esos
contenedores de palabras. En el siguiente conversatorio, algunos expertos de ámbitos que
van desde la Comunicación a la Estadística, discutirán algunos aspectos relacionados con el
análisis de esa información, técnicas, ventajas, desventajas y problemáticas asociadas, así
como la oportunidad de incorporar esta perspectiva a los proyectos investigativos.
ENTRENAMIENTO EN ANÁLISIS EPIDEMIOLÓGICOS PARA
ESTUDIANTES DE MEDICINA
José A. Betancourt*, Luis Acao Francois*, Yanira Gonzales*& Jose Félix García**
* Universidad de las Ciencias Médicas de Camagüey, Cuba.
**Universidad Juárez Autónoma de Tabasco, México.
Es necesario incrementar las habilidades relacionadas con la actividad de investigación en
estudiantes de medicina. Objetivo: entrenar alumnos de quinto año de medicina en análisis
epidemiológicos durante actividades de educación en el trabajo y de sostenibilidad.
Método: en el Tula Aguilera los estudiantes realizaron una encuesta de 14 preguntas a 69
personas enfermas de diarreas infecciosas de junio a diciembre de 2013, se contrastaron con
131 personas sanas. Recogían diversas variables demográficas, ambientales y de modos y
estilos de vida. Para cada factor determinaron (OR) con regresión logística. Se determinó
Ro y se realizó una simulación que se comparó con la usual dispersión histórica.
Resultados: los estudiantes entrenaron como se recogen datos, como se transfieren a tablas
Excel y cómo se analizan. Se observa que las personas que no tapan los desechos sólidos
tienen aproximadamente tres veces más riesgo de enfermarse (OR=2.81; IC 1.34-6.03) que
aquellas que usan algún medio para no dejarlo a cielo abierto. Se familiarizaron con el uso
del número reproductivo básico, y con las simulaciones derivadas. Conclusiones: la
estrategia de entrenamiento fue válida, sostenible, actualizada y de gran valor en la
formación de estudiantes de Medicina. En el área analizada se demostró la transmisibilidad
debida a los patrones de contactos y se mostró el efecto de las medidas de contención
establecidas.
RATIO TYPE ESTIMATION USING THE KNOWLEDGE OF THE
AUXILIARY VARIABLE FOR RANKING AND ESTIMATING
Carlos N. Bouza*, Amer I. Al-Omari**, Agustín Santiago*** & Jose M. Sautto***
*Universidad de La Habana.
**Al al-Bayt University, Faculty of Science, Department of Mathematics, Jordan.
***Universidad Autónoma de Guerrero, Mexico.
In this paper, is analyzed the behavior of ranked set sampling considering the knowledge of
the auxiliary variable. The suggested estimators are compared with their simple random
sampling counterparts. A numerical study is developed using data from a study developed
on the contamination do to burning compost from solid waste from hospitals.
A COMPARISON BETWEEN SMOOTHING AND
REGULARIZATION APPROACHES FOR MPCC.
Gemayqzel Bouza* & Ernest Quintana**
Universidad de La Habana, Cuba.
Universidad de las Villas, Cuba.
Due to the disjunctive character of the complementarity, this kind of constraints are
parameterized. Two important ways are to smooth or to regularize the restrictions leading
to equality and inequality constraints. In this paper we will compare both strategies.
BEHAVIOR OF AGRICULTURAL INTEREST RATES IN MEXICO
Ignacio Caamal Cauich, Verna Gricel Pat Fernández, Felipe Jerónimo Ascencio
Universidad Autónoma Chapingo, México.
Credits in the rural sector of Mexico come mainly from development banks (Financiera
Rural and Fideicomisos Instituidos en Relación con la Agricultura) and in a smaller
proportion of commercial banking, the development bank works with preferential interest
rates with financial intermediaries. Interest rates represent the opportunity cost of the
money and are directly related to the capitalization of borrowers. Interest rates of the
agricultural credits provided by the development banks to the banking and non-banking
financial intermediaries are around the rate of inflation, so the real interest rate is low and
can allow capitalization of intermediary credit companies. While interest rates are high for
agricultural credits provided to borrowers, i.e., the cost of credit is high for producers,
industrialists and agricultural distributors. High interest rates and the high interests of the
credits granted to borrowers, reduce the profitability of agricultural activities.
SEASONALITY OF MEXICO PERSIAN LIME PRICES
Ignacio Caamal Cauich, Verna Gricel Pat Fernández, Isaias Garcia Luis
Universidad Autónoma Chapingo, México.
Mexico produces three types of major lemons that are known in the world, which are the
Persian lime, Mexican lemon and Italian lemon. Mexican lime production has been mainly
oriented to the domestic market and the production of Persian lime towards the foreign
market and Mexico is the biggest exporter of the fruit around the world. Persian lime
production is seasonal, so prices are changing according to the levels of production, high
production and low prices period is from April to July, average production and average
prices period is from August to November and the low production and high prices period is
from December to March. The work concentrates on the analysis of the seasonal nature of
the production and seasonality of prices of the Persian lime in Mexico. The results reflect
seasonality of Persian lime prices and the forecasts of the behavior of prices of this product
are made with de-seasonalization of prices. In general, we see that the average price of
Persian lime has increased during recent years and the ranges of variation are four times,
approximately.
CÁLCULOS DE LOS TAMAÑOS DE PRUEBA Y NIVELES DE
SIGNIFICANCIA AJUSTADOS PARA LA PRUEBA DE
BLACKWELDER
José Juan Castro Alva*, Hortensia Reyes Cervantes*, Félix Almendra Arao** &
Francisco Solano Tajonar Sanabria*
*Facultad de Ciencias Físico Matemáticas. BUAP, México.
**UPIITA del Instituto Politécnico Nacional, México.
Las pruebas de no inferioridad son procedimientos estadísticos utilizados principalmente en
el contexto de ensayos clínicos, ´estas son empleadas cuando se desea comparar la eficacia
de un tratamiento nuevo contra uno ya existente, que comúnmente es llamado tratamiento
estándar o de control. Estas pruebas se emplean con el objetivo de determinar si el
tratamiento nuevo es superior, igual o inferior, por un margen generalmente pequeño al
tratamiento estándar. En el contexto de ensayos clínicos, generalmente se supone que el
tratamiento nuevo presenta alguna ventaja sobre el tratamiento de control, por ejemplo que
el tratamiento nuevo sea más barato, de más fácil aplicación o tener menos
contraindicaciones. Entre las pruebas de no inferioridad para la comparación de dos
proporciones, la prueba de no inferioridad clásica asintótica o de Blackwelder tiene un
papel destacado debido a que es la más usada en la práctica. Debido a que la prueba de
Blackwelder es una prueba asintótica, ´esta tiene la problemática de que no respeta los
niveles de significancia nominal para el cual fue creada, por esta razón, en este trabajo se
realizan los cálculos de los tamaños de prueba reales y además se calculan los valores de
los niveles de significancia nominal ajustados, de tal forma que los tamaños de la prueba
sean menores o igual es que un nivel α dado. Estos cálculos se realizaron para diseños
balanceados (tamaños de muestra iguales n1 = n2) y para diseños desbalanceados (n1, 1.5n1)
y (1.5n2, n2) mediante un programa escrito en C + +.
DISTRIBUTION AND REDISTRIBUTION IN A DOWNTON ABBEY
ECONOMY
Irwin L. Collier
Freie Universität Berlin, Germany.
A “Downton Abbey” economy consists of two social classes—a class of landowners and a
class of workers. A single good, corn, is produced for consumption by either class or for
export to finance the import of luxury goods bought by landowners. There are few enough
landowners so as to afford themselves lives of complete leisure. Landowners use their cornrents to pay the wages of their servants or for luxury goods imported from the world
beyond Downton Abbey. Workers are indifferent between working as hired hands in
agriculture or providing domestic service. Their utility depends solely upon their
consumption of corn and time not working for pay. A simple neoclassical specification of
production and consumption in competitive markets incorporates this strict classical
separation of workers from owners to determine the aggregate equilibrium allocations of
time between domestic service, productive work and the leisure enjoyed by the workers as
well as the functional distribution of income in such a Downton Abbey economy. The
comparative statics of the model are derived and interpreted. A minimum wage combined
with work-sharing can be introduced to increase workers’ welfare and reduce inequality.
LAS TRANSFORMADAS DE WATERSHED Y WATERFALL PARA
LA SEGMENTACIÓN DE IMÁGENES DE LA ADUANA”
Jorge Del Risco Martínez* & Marta Lourdes Baguer Díaz-Romañach**
* INSTEC, Cuba.
**Universidad de La Habana, Cuba.
El presente trabajo aborda el problema de la segmentación de imágenes de escáneres de la
aduana desde la mirada de la Morfología Matemática. Como primer acercamiento al
problema más complejo de tratar con imágenes con objetos superpuestos, se presenta una
combinación interesante de las transformadas de Watershed y Waterfall que permiten
lograr resultados aceptables teniendo como objetivo captar la mayor cantidad de objetos
visibles posibles. Se presentan las implementaciones de ambas transformadas con una
estrategia para seleccionar los marcadores automáticamente y P-Algoritmo. Se muestran los
resultados de la experimentación realizada.
PROMEDIO BAYESIANO DE MODELOS EN SALUD: UNA
REVISIÓN
Lucio Díaz *, Vivian Sistachs **, Dante Covarrubias *
*
Universidad Autónoma de Guerrero, México.
**
Universidad de La Habana, Cuba.
La aplicación de los modelos de regresión logística en salud es muy conocida, el objetivo es
elegir de entre un conjunto de modelos candidatos aquel que mejor se ajuste a los datos. La
selección de modelos es un aspecto que está muy relacionado con esto. En este trabajo de
hace una revisión del Promedio Bayesiano de Modelo (BMA), primero se hace una
introducción a las bases estadísticas del BMA en regresión logística, se trata los métodos
para calcular la probabilidad a posterior del modelo, incluyendo Markov Chain Monte
Carlo (MCMC). Además se hace una revisión de algunas aplicaciones usando el Promedio
Bayesiano de Modelos en salud para comparar el mismo usando distribuciones a prioris no
informativas con informativas (g-prior, propuestas por Hanson, 2014) para los parámetros
del modelo.
STATISTICAL ANALYSIS IN THE EVALUATION OF INTENSITY
OF KNOWLEDGE WORK
Claudia Echevarría Molina*, Roberto Hernández Apaulaza**, Rosario Garza Ríos*
*
Instituto Superior Politécnico José Antonio Echeverria, Cujae, Cuba.
** Centro de Inmunología Molecular, Cuba.
The great development achieved by the computing and communications provides
organizations with extensive possibilities for automating processes and for the capture,
storage, transmission and analysis of data and information. This creates a universal trend of
increased knowledge work, which simultaneously causes increased cognitive complexity of
the worker. This work was developed into a center belonging to the biopharmaceutical
industry; in order to evaluate the intensity of knowledge work processes Formulation and
Purification of a product. The method was applied to determine the intensity level of
knowledge work in different processes and subsequently variability between the answers of
the experts through statistical methods are studied. Obtaining "cluster workers", so that is
able to identify the specific features of each cluster by characterizing it belongs is also
proposed. The cluster analysis allowed further analyze the characteristics of each group and
according to this finding better ways of organizing jobs.
ANALISIS DE DESEMPEÑO ACADEMICO DE LOS
ESTUDIANTES EN LA ASIGNATURA DE ESTADISTICA EN LA
UNIVERSIDAD PEDAGOGICA DE BENGUELA – REPÚBLICA DE
ANGOLA
Abilio Eduardo* & Carlos Bouza**
*Universidad Pedagógica de Benguela, Angola.
** Universidad de la Habana, Cuba.
En el desarrollo histórico de la actividad docente, prevalece la evaluación del cumplimiento
de los objetivos programados, del desempeño profesional dos docentes, de las actividades
evaluativas para el estudiante en el aula, en tareas, pero no siempre se valorizan los hechos
sociales de esta actividad. Las recientes investigaciones pedagógicas han permitido obtener
vías para el perfeccionamiento del proceso docente-educativo y la evaluación de su calidad.
No siempre este proceso se aprovecha óptimamente en cada centro de estudios, por lo que
resulta verdaderamente substancioso conocer en qué medida es o no efectivo, por tanto la
evaluación es una práctica tradicional en esta esfera. En este contexto, la presente
investigación establece como se diferencian los estudiantes de los distintos sistemas en un
seguimiento en las carreras a partir de su desempeño en matemáticas. Este enfoque es una
primera aproximación a ofrecer una respuesta al complejo problema científico relacionado
con el desempeño de los estudiantes y de la necesidad de mejorar el éxito escolar
INFLUENCIA DEL USO DE MEDIDAS DE
SIMILARIDAD SEMANTICA PARA EL PROCESO DE MINERIA DE
OPINION EN TWITTER
Suilan Estévez-Velarde, Alejandro Piad-Morffis, Yudivián Almeida-Cruz & Claudia
Paredes-Plasencia
Universidad de La Habana, Cuba.
La minería de opinión en Twitter es un campo de investigación de reciente auge. Sus
aplicaciones incluyen estudios de mercado, seguimiento de noticias, análisis de procesos
sociales, entre otros. La información presente en Twitter tiene características que dificultan
los procesos de aprendizaje automático. Los documentos (tweets) son cortos, y sujetos a
modificaciones y errores lingüísticos. Para aliviar estos problemas se emplean medidas de
similaridad semántica, extraídas de corpus adicionales utilizando la co-ocurrencia de los
términos. Este artículo evalúa tres medidas de similaridad construidas a partir de una
selección de artículos de Wikipedia. Las medidas se comparan en cuanto a la mejora que
producen en la clasificación de mensajes de Twitter en objetivos y subjetivos.
UNA ESTRATEGIA POR VECINDADES PARA EL PROBLEMA DE
ENRUTAMIENTO DE VEHICULOS CON COMPARTIMENTOS
Alina Fernández Arias & Laritza Limia Fernández
Universidad de La Habana, Cuba.
El Problema de Enrutamiento de Vehículos con Compartimientos (VRPC) pertenece a la
familia de problemas de enrutamiento de vehículos con restricciones de capacidad en la
flota. Dado un conjunto de n clientes con demandas conocidas de diferentes tipos de
productos m y una flota homogénea de v vehículos, donde cada vehículo posee m
compartimientos iguales tal que el compartimiento p de cada vehículo está destinado a
almacenar el producto p; el VRPC consiste en encontrar el conjunto óptimo de rutas que
permita satisfacer las demandas de todos los clientes. Como condicionantes del problema,
todas las rutas comienzan y terminan en el depósito central y cada cliente es visitado una
única vez. Este trabajo propone una estrategia heurística por penalización basada en
metaheurísticas por vecindades, en la que a partir de una solución inicial, que puede ser
factible o no, se intenta encontrar una solución óptima. En el esquema presentado se
emplean las metaheurísticas Recocido Simulado y Búsqueda de Entorno Variable
Descendente, con diferentes funciones de vecindad
HERRAMIENTA COMPUTACIONAL PARA LA PLANIFICACIÓN
DE RUTAS EN EL PROBLEMA DE ENRUTAMIENTO DE
VEHÍCULOS CONSIDERANDO UNA FLOTA HETEROGÉNEA
Dafne García de Armas & Alina Fernández Arias
Universidad de La Habana, Cuba.
El problema de enrutamiento de vehículos considerando una flota heterogénea consiste en
diseñar un sistema de rutas que permita satisfacer la demanda de entrega de mercancías de
un conjunto de clientes, de forma tal que cada cliente sea atendido por un único vehículo en
una única visita. La particularidad de este problema es que la flota está compuesta por
vehículos diferentes en cuanto a capacidad y costo asociados. Atendiendo a que encontrar
una solución factible, cuando la flota posee un número limitado de vehículos, es un
problema NP-duro, en este trabajo se presenta una estrategia de penalización basada en la
Metaheurística Recocido Simulado con diferentes estrategias de vecindad. Dado que la
factibilidad de las rutas no depende del orden en el que fueron visitados los clientes, se
intenta reorganizar cada ruta según el recorrido óptimo del viajante, aplicando un algoritmo
de ramificación y acotación, cuando el número de clientes es relativamente pequeño o un
algoritmo genético en otro caso. Se diseña una herramienta computacional que brinda los
resultados para las diferentes variantes del algoritmo implementado y permite visualizar
gráficamente las rutas creadas por la solución. Es posible también, comparar los resultados
en las distintas fases de la estrategia creada.
POBREZA Y OTROS DETERMINANTES SOCIO-ECONÓMICOS Y
FINANCIEROS DEL ACCESO A VIVIENDA DE INTERÉS
PRIORITARIO EN AMÉRICA LATINA. EL CASO COLOMBIANO
José Félix García Rodríguez *, Paola Carvajal Calderón**, Alexander Cotte
Poveda**,***
*Universidad Juárez Autónoma de Tabasco, México.
**Universidad de La Salle, Bogotá, Colombia.
***Universidad de Göttingen, Alemania.
La pobreza y carencia de ingresos en los hogares, junto con otras variables
socioeconómicas y financieras, constituye el principal obstáculo para que la población
latinoamericana pueda acceder a una vivienda propia. Este estudio tiene como principal
propósito analizar los diversos elementos que influyen en la probabilidad que tienen los
hogares de bajos ingresos de contar con vivienda en América Latina, tomando como
referencia el caso de Colombia. De manera específica, se busca establecer los efectos e
incidencia de las variables asociadas a factores económicos, demográficos, sociales y
financieros. El análisis se lleva a cabo mediante la aplicación de modelos de elección
discreta logit y probit. Los resultados de los modelos probabilísticos estimados muestran
que los hogares con ingresos y niveles adecuados de ahorro tienen una mayor probabilidad
de acceder a vivienda de interés prioritario, mientras que en hogares con mayor número de
miembros, más edad y elevados pagos en el arriendo, es menor la probabilidad de obtener
vivienda. Los principales hallazgos manifiestan la importancia del diseño de estrategias y
políticas públicas que favorezcan la igualdad en la distribución del ingreso y reduzcan la
pobreza, factores claves para impulsar el crecimiento y el desarrollo económico. Como
recomendación de política de vivienda, se propone mejorar las instancias de fomento del
ahorro para vivienda.
EMPLOYMENT OF NEW GRADUATES IN THE MEXICAN LABOR
MARKET (CASE STUDY, UNIVERSITY OF THE STATE OF
MEXICO)
Gabriela Gaviño, Javier Lapa, Juan Carlos Baltazar & Ángel Sánchez
CU UAEM Valle de México, Mexico.
Weak growth of the Mexican economy in the last thirty years has created many social
problems, among which stands out the insufficient generation of formal jobs in relation to
the increasing trend the Economically Active Population (EAP). Lack of employment is
one of the most serious social problems, given its close relationship with poverty levels in a
country. In this sense the evolution of employment and unemployment in Mexico is
analyzed as of the neo-liberal economic reforms, implemented in the late eighties.
In particular, employment levels by academic degree are analyzed with the objective to
know, on the one hand, the level of interference of university studies in obtaining formal
employment in the current context, and on the other, if the new economic model has
promoted greater specialization in employment sources available or not. To do this, a case
study is developed, based on the graduates of the Autonomous University of the State of
Mexico. In this, the professional performance of these graduates, in terms of obtaining
formal employment and income obtained is evaluated. The results are compared with the
national average.
CALCULATE CHARACTERIZATION OF MONOPSONIC DEGREE
INTO THE WASTE SOLID MARKET IN LARGE METROPOLITAN
OF BRAZIL
Rilton Gonçalo B. Primo* & José Félix García Rodríguez**
*Universidad de la Habana.
**Universidad Juárez Autónoma de Tabasco.
The goal of this assay is encouragement any contribute to calculated characterization of
monopsonic degree into the waste solid market in large metropolitan of Brazil, estimating
the corresponding deviations between prices and values added, privileging cyclical and
structurally the links in the chain of reverse logistics of the residue from the scavenger
loose to industry. Makes a theoretical approach of the social significance of concentrated
markets and concludes by presenting a tool for calculating the degree of monopoly, first
steep to Fair Trade.
THE THEORY OF SEQUENTIAL LOTTERIES
Nava Kahana* ** & Doron Klunover*
*Department of Economics, Bar-Ilan University, 52900 Ramat-Gan, Israel.
**IZA, Germany.
In a sequential lottery contest with players the choice of a player equilibrium
expenditure is equivalent to choosing aggregate equilibrium expenditure in the
contest. Using that equivalence we prove that in that contest: 1. exist a unique
subgame perfect equilibrium in pure strategies and equilibrium expenditures are
homogenous in degree one in the value of the prize 2. Aggregate expenditure is
monotonically increasing in but in the limit do not reach the value of the prize and
in large contests, aggregate expenditure is lower than in a simultaneous lottery
contest.
PONTRYAGIN MAXIMUM PRINCIPLE IN OPTIMAL CONTROL
WITH DELAY IN STATE SPACE
Mamadou I. Koné
Universite Paris 1, France.
The Pontryagin principle have been introduced by Lev Pontryagin et al. in their book
Mathematical Theory of Optimal Process. They solved the following problem:
min 𝐽(𝑥(∙), 𝑢(∙)) ≔ 𝑔0 (𝑥(𝑇))
(1)
, (𝑡)
Subject to: 𝑥
= 𝑓(𝑡, 𝑥(𝑡), 𝑢(𝑡)), ∀𝑡 ∈ [0,1]
(2)
𝑥(0) = 
(3)
𝑗 (𝑥(𝑇))
∀𝑗 = 1; … ; 𝑝, 𝑔
≥0
(4)
,
𝑗 (𝑥(𝑇))
∀𝑗 = 𝑝 + 1; … ; 𝑝 + 𝑝 , 𝑔
≥0
(5)
In our works we'll study the case where there is a delay in state space. This will change the
problem from finite dimensional problem to infinite dimensional one. The main change
occurs in the dynamical constraint of the optimal problem, instead of looking for state
instantaneously, we take into account the past of the state variable. In Pontryagin frame
works, the dynamical constraint is of the form:
𝑥 , (𝑡) = 𝑓(𝑡, 𝑥(𝑡), 𝑢(𝑡))
In our study we work with :
𝑥 , (𝑡) = 𝑓(𝑡, 𝑥𝑡 , 𝑢(𝑡))
(6)
In a first step we find a resolvant and an explicit solution of (6). Then we find the adjoint
equation, its resolvant and explicit solution. In the second step we produce the link between
the two resolvants. In the third step we produce the constants, the adjoint solution and the
maximum principle.
INFERENCE OF SOURCE ORIGIN IN EPIDEMICS AND WATER
DISTRIBUTION SYSTEMS
Alejandro Lage-Castellanos*, Ernesto Ortega*, Alfredo Braunstein **, Riccardo
Zecchina**, Luca Dall'Asta** & Fabrizio Altarelli**
* Universidad de la Habana, Cuba.
**Politecnico di Torino, Italy.
We study Bayesian inference problems for epidemic models on networks using tools
borrowed from statistical physics. For each node in a known network, we obtain
approximated values for the posterior probability of being the origing of the observed
epidemic. Our approach, based on belief propagation and exact on tree-like networks, is
quite general and formal (at difference with most existing method) allowing many
observation situations, including unobserved nodes, state observations made at different or
unknown times, and observations of infection times, possibly mixed together. We ran
extensive simulations of the susceptible-infected-recovered or susceptible-infected models
of epidemics, and tested our method by inferring the epidemic origin after watching the
network state at time T. Numerical simulations show that our method outperforms previous
ones on both synthetic and real networks, often by a very large margin.
We explore a similar (but deterministic) approach in a more realistic problem: the detection
of contamination origins in water distribution systems. For this more simplified case, a
good solution can be achieved by linear programming methods, at least on synthetic data.
Applications to city-like data simulated with EPANET, however, is more problematic, and
show the need to introduce stochasticity therefore resorting to the previous belief
propagation approach.
A GPU FRAMEWORKS TO COMPUTE SUPERVOXELS IN LARGE
MEDICAL IMAGES
Antoine Leblond* and Claude Kauffmann**
*CHUM - University of Montreal Hospital Centre, Qc, Canada.
**CRCHUM - University of Montreal Hospital Research Centre, Qc, Canada.
Computation of supervoxels in large medical images is often the first step to efficient organ
segmentation. In this abstract we propose an efficient GPU-based cellular automata
framework to compute supervoxels using the Bellman-Ford algorithm. However, achieving
interactive speeds (<500 ms) in large medical images remains a challenge. Such
frameworks may suffer from degraded performance that synchronisation primitives
introduce when the label and the distance accumulator may not be encoded in a single 32bit integer because of the high number of required labels with supervoxels segmentation.
Methods: We propose a novel algorithm, the Approximation /Error-Recovery (AER)
Algorithm that get rid of most of the synchronisation primitives in favor of atomic
operations. This algorithm encodes only a part of the label, enabling encoding within a
single 32-bit integer, before the cell automata are ran to convergence. The algorithm then
tries to recover the rest of the label including error detection and correction if needed.
Results: The AER Algorithm improved the computation of supervoxels by a factor of 2.02.
Computation of supervoxels for a 512x512x885 volume (seeds every 8 voxels) took 424
ms. Conclusion: With an optimized GPU based cell automata framework and the AER
Algorithm, it was possible to compute supervoxels for large medical images at interactive
speeds (<500 ms).
ESTUDIO DEL PRONÓSTICO DE LA GRAVEDAD DEL PACIENTE:
NEUMONÍAS ADQUIRIDAS EN LA COMUNIDAD. HOSPITAL
GENERAL DOCENTE IVÁN PORTUONDO, SAN ANTONIO DE LOS
BAÑOS.
Elio León*, Carlos Bouza**, Humberto Regalado*** & Martin Rapado**
*Facultad de Ciencias Médicas de Artemisa, Cuba.
**Universidad de la Habana, Cuba.
***Hospital General Docente Iván Portuondo, Cuba.
Se ha realizado un estudio observacional analítico, prospectivo, con el objetivo de evaluar
el valor pronóstico de un grupo de parámetros clínicos en la evolución del paciente con
neumonía adquirida en la comunidad, en el H.G.D, Iván Portuondo del municipio San
Antonio de los Baños, provincia Artemisa, de setiembre 2013 a diciembre 2014. A partir de
la muestra considerada generada por un mecanismo probabilístico de 123 pacientes se
observaron al momento del ingreso las variables de interés: edad, frecuencia respiratoria,
tensión arterial., complicaciones, intolerancia a la vía oral, comorbilidad, oxigenometria de
pulso, estado de conciencia, valor de glucemia, junto a otra variables clínicas de apoyo.
Como variable respuesta se ha utilizado la restitución integra del paciente (estado al egreso)
y también ingreso en la unidad de cuidados intensivos. Se ajustaron una serie de modelos
logísticos binarios para ambas variables respuesta como métodos fundamentales de
procesamiento. La bondad del ajuste de cada modelo se evaluó a través de curvas ROC y el
test de Homers-Lemshow. Como resultados preliminares puede indicarse que la respuesta
restitución, los predictores más significativos estadísticamente fueron comorbilidad y
complicaciones, mientras que con respuesta ingreso en UCI, los más significativos fueron
vía oral y SAT O2.
PROGNOSTIC FACTORS FOR CERVICAL CANCER PATIENTS IN
TABASCO, MÉXICO
Lucas López Segovia & Daniel Romero
Universidad Juarez Autónoma de Tabasco, Mexico.
We present a survival analysis of a sample of 119 patients with cervical cancer treated at
the oncology unit HJG of Villahermosa, Tabasco, Mexico. All patients are treated with
radiation therapy or chemotherapy, and were followed from diagnosis to last visit or until
the relapse or death by cancer. Patients who relapsed or died by other causes than cancer,
has a partial survival time, and were defined as patients censored. We study the disease-free
survival time (time to recurrence of cancer) and overall survival time (time to death from
cancer) as response variables. A nonparametric analysis of survival is carried out with these
data to evaluate the effectiveness of treatment and identify groups of patients with similar
survival by its cancer characteristics. A semiparametric analysis is applied to identify
statistically significant risk factors, such as, tumour type, stage, histology, treatment, age,
among other. The results are compared with those obtained with a parametric analysis.
Preliminary results show that patients with tumour stage IIIB have the highest risk for
relapse than other patients. Patients who have tumour stage IIIB or proctitis present have
the highest risk of death than rest.
ANALISIS DE CORRELACION CANONICA USANDO
ALGORITMOS GENETICOS
Brenda Catalina Matías Castillo Gladys Linares Fleites, Hortensia Josefina Reyes
Cervantes & María de Lourdes Sandoval Solís
Benemérita Universidad Autónoma de Puebla, México.
El Análisis de Correlación Canónica (ACC) es un método exploratorio de datos
multivariados. Su propósito es la exploración de las correlaciones muestrales entre dos
conjuntos de variables cuantitativas observadas sobre el mismo conjunto de unidades
experimentales, a través de combinaciones lineales de las variables iniciales que permiten
reducir la dimensionalidad. Para encontrar los valores de las correlaciones Canónicas y
variables canónicas se tiene como propuesta usar el método de multiplicadores de
Lagrange, ya que el problema de las correlaciones canónicas puede ser visto como un
problema de máximos con restricciones, posteriormente nos lleva a un problema de valores
y vectores propios. Por otro lado, existen métodos como los Algoritmos Gen éticos (AG)
que son métodos adaptativos que pueden usarse para resolver problemas de búsqueda y
optimización global. Estos métodos están basados en el proceso gen ético de los
organismos vivos. En este trabajo se tiene como propuesta resolver el problema del ACC
usando algoritmos gen éticos como una alternativa para darle solución.
CARACTERIZACIÓN DE LA DISTRIBUCIÓN DE LOS VALORES P
DE LAS PRUEBAS PARAMÉTRICAS EN POBLACIONES
NORMALES
Pedro A. Monterrey
Departamento de Matemáticas, Universidad del Rosario, Bogotá, Colombia.
Hypothesis test are widely used in data analysis but, in general, the random nature of p
value is not consider in such applications. Power function and p-value probability
distribution are strongly related, this relationship should be a component in both hypothesis
testing conclusions and sample size calculation. However, this is not the case, possibly due
of to two factors: first many people ignore the random nature of p-value and second there is
a lack of information in the subject. Some papers introduced some characteristics of p-value
distribution as mean value and median, but such values were not obtained for every
parametric test in normal populations. This paper contains a full characterization of pvalues probability distribution for both one and two sided hypothesis test for parameters in
one or two normal populations. P-value probability distribution and their percentiles are
presented, emphasizing in median value. Additionally the paper introduces approximate
criteria for calculating both distribution function and percentiles, and highlight the
relationship between power function and p-value.
A DUAL METHOD TO SOLVE A QUADRATIC PROGRAM
DERIVED FROM A DISTANCE METRIC LEARNING PROBLEM
Bac Nguyen*, Carlos Morell* & Bernard de Baets**
*Universidad Central \Marta Abreu" de Las Villas, Cuba.
**Ghent University Coupure links 653, B-9000 Gent, Belgium.
Quadratic programming is a special type of mathematical optimization where one needs to
maximize (minimize) a quadratic function of several variables subject to linear constraints
on these variables. When the number of constraints, as well as the number of variables are
huge, general propose methods become intractable. However, in some speci_c application
domain, we can take some advantages from the problem to improve the optimization
process performance. The main idea of this work is to propose a new approach to solve a
special quadratic program large scale constraints, which is formulated to solve a problem of
distance metric learning.
UN ESTUDIO ESTADÍSTICO SOBRE EL USO PROBLEMÁTICO
DEL MÓVIL Y PERSONALIDAD
Mª Angustias Olivencia-Carrión*, Francisca López-Torrecillas*, Mª Mar Rueda* & Eva
Castillo- Fernández**
*Universidad de Granada, España.
**Centro Provincial de Drogodependencias de Granada, España.
Las investigaciones realizadas han confirmado que el uso excesivo del teléfono móvil
genera efectos perjudiciales en la salud y en las relaciones sociales. El objetivo del presente
estudio estadístico es analizar las variables de personalidad asociadas al uso versus abuso
del teléfono móvil. 157 participantes fueron divididos en tres grupos en función del tiempo
de uso diario del teléfono móvil, el Grupo 1 usaba el teléfono móvil menos de media hora
al día, el grupo 2 usaba el teléfono móvil menos de dos horas al día, y el grupo 3 usaba el
teléfono móvil más de cuatro horas al día. Para comprobar las hipótesis, se realizó un
Análisis Multivariado de la Varianza (MANOVA), para un diseño factorial, utilizando las
variables Grupo [G1, G2 y G3] como variable independiente y las variables derivadas del
Cuestionario Factorial de Personalidad de Cattel (16-PF-5) como variables dependientes.
Los resultados indicaron que había diferencias significativas entre los grupos en las
variables de personalidad.
WIKIPEDIA AS A CORPUS FOR TEXT MINING
Claudia Paredes-Plasencia, Suilan Estévez-Velarde & Alejandro Piad-Morffis
University of Havana, Cuba.
Text mining usually requires information extracted from different corpora. Wikipedia is an
encyclopedia that has earned popularity as a corpus, due to the number and quality of the
articles it contains. This paper describes the general features of Wikipedia. We evaluate the
characteristics of the grammatical structure of the articles in the Spanish Wikipedia. For
each of these characteristics, we analyze its potential influence in the process of text
mining.
E-BUFFER. A GPU FRIENDLY SPACE PARTITIONING BASED ON
BOUNDING VOLUME HIERARCHY OF EMPTY-SPACES
SUITABLE FOR RAYCASTING APPLICATIONS
Alejandro Piad-Morffis & Ludwig LeonardUniversity of Havana, Cuba.
Recent advances on graphics processing hardware have enabled the implementation of
global illumination techniques with GPU acceleration. A view-dependent representation of
the scene can be helpful for achieving interactive frame rates, since it provides refining
based on the observer's perceived level of detail and can be usually constructed using
rasterization algorithms available in modern graphics hardware. The A-Buffer structure
consists of a fragment decomposition of the scene in view space, which can be used for
multi-fragment effects such as order independent transparency, CSG visualization and
volume rendering. In this paper we introduce the E-Buffer data structure as an extension of
the A-Buffer, which stores empty-space boxes in view-space to enable an adaptive ray-
marching function. We propose a GPU accelerated construction of the E-Buffer that allows
a virtually unbounded number of fragments per pixel, and a generalized ray casting
operation using the data stored in the structure. For illustrating the E-Buffer utility we
present a real-time GPU ray tracer.
HIPERENTORNO BASADO EN OBJETOS DE APRENDIZAJE
DESTINADO A LA ASIGNATURA ANÁLISIS DE DATOS (ESTADÍSTICA)
PARA LA CARRERA DE CULTURA FÍSICA.
José Antonio Pino Roque, Yasser Vázquez Alfonso, Mayra Arteaga Barrueta, Lázaro B. García
Gutiérrez, Lucía Fernández Chuairey & Josefina de Calzadilla Pereira*
UNAH, Cuba.
Partiendo de la contradicción existente entre el modelo de estudios semipresencial con apoyo de
las TIC´s y la carencia de medios interactivos y de autoevaluación con apoyo de las TIC´s con que
cuentan los Centros Universitarios Municipales (CUM) para lograr los objetivos propuestos de este
modelo de estudio, se fundamenta esta investigación, la cual diseña un hiperentorno basado en
Objetos de Aprendizaje (OA) destinado a la asignatura Análisis de Datos (Estadística) para la
carrera de Cultura Física. El resultado fundamental de esta investigación es la obtención de un
material digital con recursos propios del Proyecto de Tecnología Educativa (LATED) de la
Universidad Agraria de la Habana ¨Fructuoso Rodríguez Pérez¨, a partir de concepciones
pedagógico-tecnológicas, la cual favorece la superación del estudiante creándole una base sólida
para esta asignatura de la disciplina Métodos de Análisis e Investigación. Los resultados obtenidos
con la aplicación de este recurso digital han sido muy positivos en el proceso de enseñanza
aprendizaje en las CUM de Mayabeque y Artemisa. La utilización del Exe-learning, y las
herramientas y metodología de producción diseñadas por LATED han sido probadas con éxito
proporcionando un entorno virtual de enseñanza aprendizaje muy aceptada por estudiantes y
profesores de los CUM y la Universidad.
LOS ESCENARIOS DE LA GLOBALIZACION Y LA POBREZA EN
MEXICO
Oscar Priego Hernández, José Félix García Rodríguez & Miguel Angel Ramírez
Universidad Juárez Autónoma de Tabasco, Cuba.
México es uno de los países de la OCDE en donde se pueden apreciar los impactos de las
medidas económicas adoptadas en las últimas décadas. El modelo neoliberal, formalizado a
partir de 1987, tiene sus éxitos como reducción de la inflación, eliminación del déficit fiscal
y un moderado crecimiento económico. Sin embargo es necesario reconocer la fragilidad
de tales éxitos respecto a sus costos económicos y sociales. Esto se puede apreciar en la
desproporción que hay en el pobre crecimiento económico y calidad de vida de la
población. Hoy los escenarios de la economía mexicana se encuentran ante una gran
disyuntiva que implica buscar el crecimiento económico para generar los empleos que el
país necesita o seguir con la política de mantener controladas las principales variables
macroeconómicas, inflación y tasas de cambio. Ante este escenario, México enfrenta
además nuevos retos como continuar financiando el gasto público mediante la renta
petrolera. El verdadero reto es dejar paulatinamente la dependencia de los ingresos
petroleros y para ello se requiere de la voluntad política de todos los actores. En esta
investigación se plantean propuestas a manera de estrategias para mejorar la economía del
país que permitan el crecimiento económico y la seguridad que los mexicanos esperan en
este siglo. La metodología utilizada es de enfoque cuantitativa mediante la aplicación de
técnicas de análisis para obtener información.
LEVERAGING EXPERTS KNOWLEDGE TO IMPROVE
DIAGNOSTICS ON AIRCRAFT ENGINE
Tsirizo Rabenoro*,**, Jérôme Lacaille*, Marie Cottrell** & Fabrice Rossi**
*SNECMA, Safran Group, France.
**Paris 1 University Pantheon Sorbonne, France.
Aircraft engines are extremely reliable thanks to the manufacturing process. Thus
operational events, such as in flight shut down (IFSD) or delays and cancelation (D&C), are
rare. However, there is a need to help companies optimize maintenance costs and reduce
those operational events. Such events can lead to expensive unexpected maintenance costs.
Moreover delays and cancelations mean troubles for customers. As a turbofan engine
manufacturer, Snecma has all the knowledge required for development of efficient Health
Monitoring systems.
Large amount of engine related data are collected during flights. These data are analyzed to
detect early signs of anomalies. According to experts, specific measurements and
computation may help to identify each fault. Snecma leverages experts’ knowledge by
developing algorithms designed to automatically detect early signs of anomaly.
Algorithms are hard to calibrate. Extreme reliability of engines makes it difficult to
maintain a low level of false alarms. Thus algorithms are constantly tuned based on
experts’ knowledge and feedback. Moreover each algorithm focuses mainly on a subsystem
despite the need of a diagnostic of the whole system. Once an anomaly is detected, one
needs to identify the origin of the anomaly, but the dependencies between subsystems are
complex to model. Sometimes, a shop visit is needed to identify the origin of the problem.
In this article, we introduce a methodology that helps acquire and leverage experts’
knowledge and feedback. This methodology automatically tunes algorithms and
automatically set fusion rules in order to identify the origin of the anomaly. The idea is to
collect result of each analysis done by experts or during shop visit by using a labellisation
tool. The current method uses a very large number of binary indicators based on algorithms
designed by experts. Each indicator corresponds to the response of an algorithm with a
specific set of parameters. The binary homogenization allows us to combine different kinds
of algorithms. Then the best indicators are selected leading to a much reduced number of
indicators that are tuned on a learning data set and then are aggregated to provide an
interpretable result giving the presence of an anomaly and clues to the origin of this
anomaly. This allows us, through the selection process, to automatically calibrate
algorithms. The binary indicators allow human operator to make sense of the results
provided by the methodology. Thus a human operator can confirm results by using the
labellisation tool. And then, if a too high level of false alarm is noted, the selection and the
aggregation of binary indicators can be relaunched to improve diagnostic performance with
a new iteration of the learning procedure. Finally, Snecma will evaluate this procedure
within the context of the developments of its future projects.
MODELACION ESTADISTICA SOBRE EL EGRESO DE
UNIVERSITARIOS DE CIENCIAS EXACTAS DE LA FCFM-BUAP
Hortensia Reyes Cervantes*, Ana Luisa Nieto Méndez*, Fernando Velasco Luna* &
Flaviano Godínez Jaimes**
*Benemérita Universidad Autónoma de Puebla, México.
**Universidad Autónoma de Guerrero, México.
Los métodos de la estadística matemática dan la posibilidad de presentar el conjunto de
resultados observados en una forma completa, reducida y adecuada para su análisis. En
particular el modelo de regresión logística permite explicar el comportamiento de una
respuesta binaria mediante varias variables independientes explicativas de naturaleza
cuantitativa y/o cualitativa. La eficiencia terminal es un indicador importante en las metas y
objetivos que tienen las universidades hoy en día, aunque éstas no pueden ser primordiales
porque existen otros factores relacionados. En todas las instituciones es importante conocer
o estimar el periodo de estancia de los alumnos, ya que existe mucha deserción muy
comúnmente en las licenciaturas de ciencias: la reprobación, rezago y deserción. Se
estudiaron factores académicos para que un alumno egrese o no de alguna de las
licenciaturas de ciencias exactas que se imparten en la facultad usando regresión logística.
Las materias fueron 14 variables para las carreras del área de Matemáticas (Matemáticas y
Matemáticas Aplicadas) y 21 para las carreras del área de Física (Física y Física Aplicada).
Mediante una selección de variables, se encontraron las materias que explican el
comportamiento del egreso por área y se obtuvo que aunque ingresan más hombres que
mujeres en las licenciaturas, en la licenciatura del área de física, egresan más hombres y en
el área de Matemáticas, egresan más mujeres.
ALGORITMO GENÉTICO SEMIDIFUSO.
APLICACIÓN A UN MODELO DE DETECCIÓN DEL VIH EN
CUBA
Lianne Rodríguez Matos*, Greter Domínguez Rodríguez** & Aymée Marrero Severo**
*Instituto Superior Politécnico José Antonio Echeverría, Cuba.
**
Universidad de La Habana, Cuba.
En este trabajo se presenta el diseño de un Algoritmo Genético Semidifuso concebido para
resolver el problema de estimación de parámetros que surge al modelar el contagio de la
epidemia VIH (SIDA) en Cuba como un sistema dinámico mediante Ecuaciones
Diferenciales Ordinarias (EDO). El objetivo principal del trabajo es consolidar las bases
para el desarrollo en el campo de la creatividad computacional. Combinando el Algoritmo
Genético Tradicional (AG) con la Lógica Difusa se aspira a mejorar la diversidad
poblacional del AG de manera que se superen los problemas de convergencia prematura
exhibidos en los mismos. El diseño presentado es una adaptación de un AG en el que se
trabaja con una representación difusa de los individuos de la población y operadores de
cruzamiento y mutación también difusos, dejando sin fuzzificar solo la función de
adaptación. Para la implementación del algoritmo se utilizó C# como lenguaje de
programación, Visual Studio 2012 como entorno de desarrollo y la biblioteca ALGLIB
como soporte para los métodos numéricos.
EVALUACIÓN DE LA POBREZA, DESARROLLO COMUNITARIO O
DESARROLLO LOCAL Y EL MUNICIPIO.
Lorenzo Salgado García Alejandro Jimenez Jimenez &Maritza Valdez Rodríguez
Benemérita Universidad Autónoma de Puebla, México.
La experiencia de vinculación que se establece entre los programas de servicio social y las
Universidades con las comunidades, parten de un proyecto de investigación en el municipio
de San Buena Ventura Nealticán; en el marco del desarrollo del proyecto denominado
“laboratorio móvil” en atención multidisciplinaria a la salud en comunidades prioritarias
del Estado de Puebla. Es relevante señalar que, el grado de pobreza y marginación social
que existe en el municipio, obligó a los habitantes a emigrar hacia el país vecino del norte,
por lo tanto, las remesas se han ocupado del desarrollo económico local y de formación de
negocios de las familias.
Por ello, el objetivo del trabajo es, analizar el desarrollo de las comunidades que integran al
municipio, su estructura, explotación de recursos y grado de marginación social. De igual
forma determinar algunos factores que deberían ser atendidos por la administración
municipal en la búsqueda de un desarrollo local equitativo, que ayude a combatir la
pobreza, la migración, la marginación social entre otros problemas para el desarrollo del
municipio.
GUIDED DECISION PROCESSES
Rakesh K. Sarin
University of California, USA.
The heuristics and bias research program has convincingly demonstrated that our
judgments and choices are prone to systematic errors. Decision analysis requires coherent
judgments about beliefs (probabilities) and tastes (utilities), and a rational procedure to
combine them so that choices maximize subjective expected utility. A guided decision
process is a middle-of-the-road between decision analysis and intuitive judgments in which
the emphasis is on improving decisions through simple decision rules. These rules reduce
cost of thinking, or decision effort, for the myriad decisions one faces in daily life; but at
the same time, they are personalized to the individual and produce near optimal choices.
We discuss the principles behind the guided decision processes research program, and
illustrate the approach using several examples.
A PRACTICAL USE OF RADIAL BASIS FUNCTIONS INTERPOLATION AND
APPROXIMATION
Vaclav Skala
University of West Bohemia, Faculty of Applied Sciences, Czech Republic.
Radial basis function, RBF, interpolation, approximation, numerical methods Interpolation
and approximation methods are used across many fields. Standard interpolation and
approximation methods rely on “ordering” that actually means tessellation in d-dimensional
space in general, like sorting, triangulation, generating of tetrahedral meshes etc.
Tessellation algorithms are quite complex in d-dimensional case. On the other hand,
interpolation and approximation can be made using meshfree (meshless) techniques using
Radial Basis Function (RBF). The RBF interpolation and approximation methods lead
generally to the solution of linear system of equations 𝑨𝒙 = 𝒃. However, a similar
approach can be taken for a reconstruction of a surface of scanned objects. In this case this
leads to a system of homogeneous system of equations 𝑨𝒙 = 𝟎, when a different approach
has to be taken. In this paper we describe novel approaches based on RBFs for data
interpolation and approximation generally in d-dimensional space. We will show properties
and differences of “global” and “Compactly Supported RBF (CSRBF)”, run-time and
memory complexities. As the RBF interpolation and approximation naturally offer
smoothness, we will analyze such properties as well as approaches how to decrease
computational expenses. The proposed meshless interpolation and approximation will be
demonstrated on different problems, e.g. inpainting removal, restoration of corrupted
images with high percentage of corrupted pixels, digital terrain interpolation and
approximation for GIS applications and methods for decreasing computational complexity.
Experimental results will be presented as well.
META-ANALYSIS OF SAFETY AND EFFICACY IN PATIENTS
WITH DIABETIC FOOT ULCERS. INTERNATIONAL PROYECTION
OF HEBERPROT-P®.
Angela Tuero , et al.
Center for Genetic Engineering and Biotechnology (CIGB), Cuba.
Heberprot-P® is a novel therapeutic in patients with Diabetic Foot Ulcers (DFU), that have
been extended to some regions of the world. Objective: To evaluate the results of safety and
efficacy of the Heberprot-P outside Cuba in the period of 2010-2014, using the metaanalysis techniques.
A meta-analysis was done allowing to integrate findings, accumulate data and identify the
general tendency for efficacy variables (granulation tissue, healing and amputation) and
safety: adverse events (type, intensity, seriousness and causality relation). It was included
the results of the studies in 8 countries during the period 2010-2014. Q of Cochran test was
used to evaluate the homogeneity between them. Depending of the data assumptions,
models for fixed and random effects were adjusted. Results: Around 90% of patients had
complete granulation and the 85% reached healing. The rate of major amputation was
small. The safety profile was adequate. The more frequently adverse events (> 10%) were:
pain and burning sensation in the application site, shivering and chills. It were predominant
the events with mild and moderate intensity. There were no onco-proliferative process
reported. The application of Heberprot-P in patients with DFU outside the national region
maintain a positive relation benefit/ risk for the patients.
SIMULACIÓN PARA EVALUAR EL TRATAMIENTO DE DATOS
FALTANTES EN MODELOS DE TRANSICIÓN EN EL CONTEXTO
DE ENSAYOS CLÍNICOS
Rolando Uranga*, Sira Allende** & Geert Molenberghs***
*
Centro Nacional Coordinador de Ensayos Clínicos, Cuba.
** Universidad de la Habana, Cuba.
*** Universidad de Hasselt, Bélgica.
En estudios longitudinales no se suele contar con la completitud de los datos. En este
trabajo se comparan, mediante un estudio de simulación, distintos métodos para abordar la
estimación bajo datos faltantes en estudios longitudinales compatibles con un modelo de
transición. Se consideran mecanismos de pérdidas aleatorio de tipo abandono y
completamente aleatorio de tipo intermitente. Se incursiona en mecanismos adicionales no
aleatorios. Los métodos empleados incluyen casos completos, imputación por cero,
imputación por la media y última observación llevada adelante. Se refinan los métodos
mediante imputación por la media condicional, regresión y regresión estocástica. Tales
métodos se contrastan con los de verosimilitud directa e imputaciones múltiples. Se valora
la bondad de cada uno en términos del porciento de pérdidas, que se garantiza mediante un
algoritmo de base empírica construido al efecto. Se introduce un criterio cualitativo que
permite realizar una valoración rápida y clara de cada método, mediante su clasificación en
Excelente, Bien, Regular o Mal. Se obtiene de este modo un simulador, escrito en el
lenguaje IML (Interactive Matrix Language) de SAS (Statistical Analysis System), que
permite realizar la evaluación. Se concluye que los métodos de imputación ingenua
producen estimadores sesgados. Métodos de imputación simple refinados como regresión
estocástica, muestran comportamiento aceptable para bajos porcientos de pérdidas aunque
subestiman los parámetros de variabilidad. El método de casos completos muestra buen
comportamiento. Verosimilitud directa e imputaciones múltiples son los métodos de mejor
comportamiento.
DISEÑO DE UN PROCEDIMIENTO PARA LA PLANIFICACIÓN Y CONTROL
DE LA PRODUCCIÓN HACIENDO USO DE HERRAMIENTAS MATEMÁTICAS
Idalianys Urquiola Garcia*, Amelia Tamayo Garcia*
* Instituto Superior Politécnico “José Antonio Echeverría”, CUJAE. Cuba.
The production planning is an ongoing process designed to determine in advance decisions
to optimize the use of productive resources. It is nothing but the sales forecast for a time
horizon, the way of finding the combination of productions, stocks and global resources of
the company, which manage to fulfill the demand at its best. Manufacturing companies
demand radical changes to achieve the required quantity and quality of their productions
and respond to market needs quickly, being necessary to do so the correct choice of
planning system and production control. The practical application of either tool depends on
the classification of the production system. Currently, there is no procedure that allows the
selection of the correct mathematical tool for planning and control to be applied in
manufacturing companies. The proposed procedure allows companies through the scores
given to the implementation of a multi-criteria approach, identifying and selecting the
mathematical tool that best suits your needs; taking into account the characteristics of the
production system, the planning horizon and its classification according to the way of
managing the flow of production. Among the proposed planning tools can be counted:
Material Requirements Planning (MRP and MRP II), Hierarchical Production Planning
(HPP), Linear Programming, Multiobjective Optimization, Information Cards (Kanban)
Constant Work in Process (Conwip), Models and Process Simulation.
CONTRAST ENHACEMENT FOR DIGITAL MAMMOGRAPHY
IMAGES BASED ON WAVELETS
Damián Valdés Santiago, Daniel Mesejo León, Ángela León Mecías
*
Universidad de La Habana, Cuba.
For Cuban women the highest rate of cancer mortality correspond to the breast cancers. 1%
of breast cancers occur in men. The most effective tool for reducing mortality is prevention
through early diagnosis to the study of asymptomatic women. As benefits of
mammography screening a reduction of 25 % is observed in the mortality rate from breast
cancer. The detection of micro calcifications (usually small and very white dots) in the
breast may be an early cancer or further development of the same sign. Mammography can
go unnoticed between 10 % and 25 % of tumors, among other factors due to poor image
quality due to low radiation. This paper aims to study the effect of wavelet based
algorithms to improve image contrast digital mammography, e.g. simple, threshold,
correlation and local correlation algorithms.
TIEMPO AL DETERIORO DE LOS SCORES DE CALIDAD DE VIDA
COMO UNA MODALIDAD DE ANÁLISIS LONGITUDINAL PARA
PACIENTES CON CÁNCER DE CABEZA Y CUELLO TRATADOS
CON NIMOTUZUMAB.
Carmen Viada et al.
Centro de Ingeniería Molecular, Cuba.
El Centro de Inmunología Molecular (CIM) es uno de los centros del Polo Científico de
Cuba dedicado a la investigación, desarrollo, producción y comercialización de productos
biotecnológicos humanos. Nimotuzumab es un anticuerpo monoclonal que reconoce el
receptor del factor de crecimiento epidérmico con una alta afinidad para el tratamiento del
cáncer, y representan una clase prometedora de nuevos fármacos contra el cáncer. En este
tipo de tratamientos es importante contar con información acerca de la percepción de la
calidad de vida que el propio paciente tiene respecto de su tratamiento como complemento
a indicadores clínicos de supervivencia en la evaluación de eficacia.
Objetivo: evaluar el tiempo al deterioro de los score de calidad de vida QLQ-C30
(instrumento genérico para cáncer) y QLQ-HN35 (instrumento específico para cáncer de
cabeza y cuello).
Material y métodos: Se evaluaron las respuestas a las encuestas de calidad de vida QLQC30 y QLQ- HN35 de 237 pacientes diagnosticadas con cáncer de cérvix incluidas en un
ensayo clínico con el nimotuzumab un receptor del factor de crecimiento epidérmico
bloqueado terapia. Se estimó el tiempo para el 5%, 10%, 20% y 30% de deterioro de la
calidad de vida con respecto a su valor basal para los 15 scores del QLQ-C30 y los 18
scores del QLQ-HN35 propuestos por la EORTC.
Resultados y Conclusiones: El tiempo al deterioro resultó ser superior para el grupo
experimental comparado con el grupo control. Los pacientes tratados con Nimotuzumab
además de tener una mayor supervivencia, tienen mejor calidad de vida.
LANGUAGE RIGHTS: A WELFARE-ECONOMICS APPROACH
Bengt-Arne Wickström
Andrássy-Universität Budapest, Hungary.
We analyze normatively determined distributions of language rights in multilingual
settings. A concave cost structure in the number of beneficiaries of providing rights leads to
a critical-mass criterion for determining the optimal rights structure. It is further shown in a
welfare-maximizing model where rights influence the status of a language, which in turn
influences the preferences for language rights that the “naive” cost-benefit analysis has to
be augmented in various directions. In an efficiency analysis, this has its roots in the intergenerational transfer of language use and the endogeneity of preferences. In a welfaremaximizing approach, the redistribution goals further modify the analysis.
POPULATION SAMPLING FUZZY
John Jairo Zabala Corrales & Dicleny Castro Carvajal
University of Tolima – Colombia.
The theory of fuzzy sets, introduced by Zadeh (1965), has become a good alternative as it
concerns the modeling world, this based on the idea that the key elements of human
thinking are not numbers but labels language that allow objects to pass to belong flexibly
from one class to another, allowing you to have multiple possibilities of interpretation.
Initially the basic concepts of fuzzy logic arise, then a presentation on populations and
samples is diffuse, then the way is shown as the arithmetic mean and all parameters are
estimated in simple random sampling, in order to compare the way to estimate in diffuse
samples fuzzy finite populations, this using indices Kaufmann (1975) as estimators.
Plenary Lecturers
Michael Ahlheim has been professor of
Environmental Economics, Regulatory and Consumer
Policy at the Institute of Economics of the University
of Hohenheim in Stuttgart since 2001. Before that, he
was professor of Environmental Economics at the
Brandenburg Technical University in Cottbus. He
holds a doctoral degree in economics (Dr. rer. pol.)
and habilitation degree (Dr. rer. pol. habil.) from the
University of Heidelberg.
His research covers different topics from
environmental economics, but also from regulatory
and consumer policy. His main field of research is the
economic valuation of environmental changes with a special focus on contingent valuation and
choice experiments. As a member of the Collaborative Research Center (Sonderforschungsbereich)
No. 565 of the German Science Foundation (DFG) on "Development and Valuation of Disturbed
Landscapes" (2001-2003) and the International Collaborative Research Center No. 564 of the
German Science Foundation on "Sustainable Land Use and Rural Development in Mountainous
Regions of Southeast Asia" (2003-2012) as well as of several interdisciplinary research groups
financed by the German Federal Ministry of Education and Research (BMBF) he conducted
empirical environmental valuation studies in different countries like China, Thailand, Vietnam, the
Philippines and Germany.
Prof. Dr. Michael Ahlheim
Institute of Economics
University of Hohenheim, Germany
James J. Cochran is Professor of Applied Statistics and the Rogers-Spivey
Faculty Fellow in the Department of Information Systems, Statistics, and
Management Science of the University of Alabama’s Culverhouse College
of Commerce and Business Administration. He earned a PhD in Statistics
from the University of Cincinnati and has been a Visiting Scholar with
Stanford University, University of South Africa, Universidad de Talca, and
Pôle Universitaire Léonard De Vinci.
Professor Cochran’s research interests include sample based and
Bayesian optimization, statistical methods (particularly general linear
models), statistical learning, and applications to sports. Professor Cochran
has served as a consultant for many companies, government agencies,
Culverhouse College of
Commerce & Business
and NPOs. He co-founded Statistics Without Borders, and he established
Administration
and has organized INFORMS’ Teaching Effectiveness Colloquium series
University of Alabama, USA
and annual Case Competition as well as an International Education
Workshop series which has taken place in Uruguay, South Africa, Colombia, India, Argentina,
Kenya, Cameroon, and Croatia. He was the General Chair of the 2005 INFORMS Conference and
organized and chaired the 2008 ORPA Conference on Using Operations Research to Address Urban
Prof. Dr. James J. Cochran
Transport and Water Resource Management Issues in Africa. Professor Cochran is founding Editorin-Chief of the Wiley Encyclopedia of Operations Research and the Management Sciences and the
Wiley Series in Operations Research and Management Science. He was 2007-2012 Editor-in-Chief
of INFORMS Transactions on Education and he currently serves on the editorial boards for several
journals.
In 2006 Professor Cochran was elected to the International Statistics Institute, in 2008 he received
the INFORMS Prize for the Teaching of OR/MS Practice, in 2010 he received the Mu Sigma Rho
Statistical Education Award, and in 2011 he was named a Fellow of the American Statistical
Association. In 2014 the American Statistical Association presented Professor Cochran with its
Founders Award, the highest honor bestowed by the ASA.
Marcos José Negreiros Gomes, is Associate professor at
Universidade Estadual do Ceará (UECE). He obtained his
doctoral degree at Universidade Federal do Rio de Janeiro. He
had a Post-Doctoral fellowship at Université de Montreal. Prof.
Negreiros coordinates the Professional Master in Science course
MCOMP-UECE/IFCE. He is a CNPq researcher who focusses in
mathematical programming for combinatorial problems, parallel
programming, graph theory, algorithms, data structures, and
GIS. In 2005 he obtained the IFORS Prize in Development. He
works with systems to integrate information for dengue disease
Prof. Marcos José Negreiros Gomes,
Universidade Estadual do Ceará (UECE
Brazil
Dr. Joseph Rynkiewicz is
Associate Professor
Université Paris-I
Member of the SAMM Research
Center, France
and control, health systems, garbage collection, planning and
scheduling multi-projects, parallel optimization, dynamic graph
problems, routing and location. He is the coordinator of the
CAPES-MÊS Cuba Project.
Joseph Rynkiewicz is associate professor at the Université Paris-I
and member of the SAMM research center.
Dr Rynkiewicz's research focuses on asymptotic properties of
statistical estimator as well as practical applications of new
statistical tools like artificial neural networks, hidden Markov
models and Bayesian networks.
His publications range from asymptotic studies of models under loss
of identifiability to forecast of ozone peaks and exceedance levels
using neural classifiers and weather predictions. He also worked for
industrial applications like speed of wing prediction for wind farm
project.
He teaches theoretical statistics, data mining, survey and computing
from license to Master at the Université Paris-I. He joined this
University after earning his PhD in applied mathematics in 2000.
Main Lecturers
L. Robin Keller is the 2015 President of the Institute for
Operations Research and the Management Sciences (INFORMS,
www.informs.org). Prof. Keller’s research focuses on developing
techniques for analyzing decisions. Her decision analysis
research spans the areas of multiple attribute decision making,
fairness, perceived risk, and planning protection against
terrorism, environmental, health, and safety risks. She is a Full
Professor of Operations and Decision Technologies in the Paul
Merage School of Business at the University of California, Irvine,
USA. She teaches management science to Health Care
Executive MBAs, decision analysis to MBAs, and decision theory
to doctoral students. She joined UCI in 1982 after earning her
M.B.A. and Ph.D. in management science from UCLA. She has
Prof. Dr. L. Robin Keller
served many roles in the school, including the Doctoral Program
Paul Merage School of Business
Director and Associate Dean (for the Full-time MBA Program
University of California, USA
and for Research). She was Program Director for the NSF
Decision, Risk, and Management Science Program. Dr. Keller was the Editor-in-Chief of Decision
Analysis from 2007-2012, a founding Director-at-large of INFORMS, TIMS Vice President-Finance
and Council member, and Decision Analysis Society Chair. She became an INFORMS Fellow in 2004
and received the 2006 Kimball Medal from INFORMS.
Mauro Oliveira
Mauro Oliveira is Professor at the Instituto Federal de Educação Ciência e Tecnologia do Ceará
(IFCE), in (Brazil), since 1974, and Professor of the Profissional Master of Science Program at the
Universidade Federal do Ceará (UECE), in Brazil. He was a Guest Professor of the DEA Program at
the Université de Technologie de Troyes (2003). He also gave lectures in the Département
D’Informatique at the Université Claude-Bernard, Lyon 1 (1991).
Born on 4 September 1954, Mauro Oliveira holds a M.Sc in Electrical Engineering from the
Universidade Católica do Rio de Janeiro (PUC-Rio) and a Ph.D in Computer Science from the
Université Pierre et Marie Curie (Paris VI, France). He then carried out a postdoctoral stage at the
Centre for Telecommunication Research (CTR) at the King’s College of London (2003).
Recently, he finished another postdoctoral stage at the School of Information Technology and
Engineering (SITE), University of Ottawa (Canadá).He participated of the first meeting (1988) for
the implementation of the Internet in Brazil and was member of the CGI.BR (Manager Committee
of the Internet Brazil). He was also chairman of the XIV SBRC-Brazilian Symposium of Computer
Networks (1996), of the XX SEMISH Brazilian Symposium of Hardware and Software (1999) and of
other important conferences in Brazil
Fátima Sombra
Universidad Federal de Ceara.
List of Participants
Agüero Zardón, L. (Cuba): Aplicación del método CDS secuencial para el rediseño de las
rutas de distribución en una fábrica productora de helado.
Ahlheim, M. (Germany): The economic valuation of changes in land use towards more
sustainability – an empirical example from southwest China.
Allende, S. (Cuba): Simulación para evaluar el tratamiento de datos faltantes en modelos
de transición en el contexto de ensayos clínicos.
Allende, S.(Cuba):Teaching OR in different contexts.
António, L. A. (Angola): Participant.
Arcos, A. (Spain): The use of landline and cell-phone in health surveys.
Arcos, M del M. (Spain): The use of landline and cell-phone in health surveys.
Arcos, M del M. (Spain): Gender analysis of medical professionals in Spanish public
universities.
Arguello Ortíz, Á. F. (Mexico): Conversatorio Internacional sobre el Análisis de Grandes
Volúmenes de Datos Textuales.
Baguer, M.L. (Cuba): Las transformadas de Watershed y Waterfall para la segmentación de
imágenes de la aduana.
Baltazar, J.C. (Mexico):Employment of new graduates in the Mexican labor market
(Case Study, University of the State of Mexico).
Baltazar, J.C. (Mexico): Econometric & statistical response model applied to socioeconomical careers at uaem. C. U. Valle de México.
Betancourt Bethencourt, J. (Cuba): Entrenamiento en análisis epidemiológicos para
estudiantes de Medicina.
Bouza G., (Cuba): A comparison between smoothing and regularization approaches for
MPCC.
Bouza, C. (Cuba): Ratio type estimation using the knowledge of the auxiliary variable for
ranking and estimating.
Caamal Cauich. I. (Mexico): Behavior of agricultural interest rates in Mexico.
Caamal Cauich. I. (Mexico): Seasonality of Mexico Persian lime prices.
Castro Alva, J.J. (Mexico): Cálculos de los tamaños de prueba y niveles de significancia
ajustados para la prueba de Blackwelder.
Castro Carvajal, D. (Colombia): Population sampling fuzzy.
Cochran, J. J (USA): Active learning: reengaging and reenergizing your students and their
instructor!
Cochran, J. J (USA): The education generation gap: confronting and embracing the
challenges that growing computing power and new technology present to applied
mathematics education.
Collier, I. L. (Germany): Distribution and redistribution in a downton abbey economy.
Díaz, L. (Mexico).:Promedio Bayesiano de Modelos en Salud: una revisión
Echevarria Molina, C. (Cuba): Statistical analysis in the evaluation of intensity of
knowledge work.
Eduardo, A, (Angola): Análisis de desempeño académico de los estudiantes en la
asignatura de estadística en la Universidad Pedagógica de Benguela – República de
Angola.
Estévez-Velarde, S. (Cuba): Wikipedia as a corpus for text mining.
Estévez-Velarde, S. (Cuba): Influence en twitter: a proposal based on clustering.
Estévez-Velarde, S. (Cuba): Influencia del uso de medidas de similaridad semántica para el
proceso de minería de opinión en Twitter.
García de Armas, D. (Cuba): Herramienta computacional para la planificación de rutas en el
problema de enrutamiento de vehículos considerando una flota heterogénea.
García Rodríguez, J. F. (Mexico): Calculate characterization of monopsonic degree into the
waste solid market in large metropolitan of Brazil.
García Rodríguez, J. F. (Mexico): Los escenarios de la globalización y la pobreza en
México.
Garcia, J. F. (Mexico): Entrenamiento en análisis epidemiológicos para estudiantes de
Medicina.
Garcia, J.F. (Mexico): Pobreza y otros determinantes socio-económicos y financieros del
acceso a vivienda de interés prioritario en América Latina. El caso colombiano.
Gaviño, G. (Mexico): Employment of new graduates in the Mexican labor market (Case
Study, University of the State of Mexico).
Gaviño, G. (Mexico): Econometric & statistical response model applied to socioeconomical careers at UAEM. C. U. Valle de México.
Gonçalo B. Primo, Rilton Brazil: Calculate characterization of monopsonic degree into the
waste solid market in large metropolitan of Brazil.
Hernández, S. (Mexico): Conversatorio Internacional sobre el Análisis de Grandes
Volúmenes de Datos Textuales.
Hillman, A. (Germany): Participant.
Jay, S. (USA): Teaching multi-objective multi-stakeholder decision modeling with cases.
Kauffmann, C. (Canada): A GPU frameworks to compute supervoxels in large medical
images.
Keller, L.R. (USA): Teaching multi-objective multi-stakeholder decision modeling with
cases.
Klunover, D. (Germany): The theory of sequential lotteries.
Koné, M.I.(France): Pontryagin maximum principle in optimal control with delay in state
space.
Laborde, M (Mexico): Conversatorio Internacional sobre el Análisis de Grandes Volúmenes
de Datos Textuales.
Lage-Castellanos, A. (Cuba): Inference of source origin in epidemics and water distribution
systems.
Lapa, J. (Mexico): Employment of new graduates in the Mexican labor market (Case Study,
University of the State of Mexico).
León Mecías, A. (Cuba): Movement detection using wavelet.
Leon, A. (Cuba): Contrast enhacement for digital mammography images based on
wavelets.
Limia, L. (Cuba): Una estrategia por vecindades para el problema de enrutamiento de
vehiculos con compartimentos
López Segovia, L. (Mexico): Prognostic factors for cervical cancer patients in Tabasco,
México
Marrero, A. (Cuba): Algoritmo genético semidifuso. Aplicación a un modelo de detección
del VIH en Cuba
Matías Castillo, B.C. (Mexico): Análisis de correlación canónica usando algoritmos
genéticos
Mesejo León, D. (Cuba): Contrast enhacement for digital mammography images based on
wavelets.
Monterrey, P.(Colombia): Caracterización de la distribución de los valores p de las
pruebas paramétricas en poblaciones normales.
Negreiros, N. (Brazil): Capacitated centered clustering problem: applications, models, exact
and metaheuristic methods.
Oliveira, M. (Brazil): Evolving an intelligent framework for decision- making process in ehealth systems.
Oller Alonso, M. (Ecuador): Conversatorio Internacional sobre el Análisis de Grandes
Volúmenes de Datos Textuales.
Piad-Morffis, A. (Cuba): Wikipedia as a corpus for text mining.
Piad-Morffis, A. (Cuba): E-buffer. A GPU friendly space partitioning based on bounding
volume hierarchy of empty-spaces suitable for raycasting applications.
Priego Hernández, O. (Mexico): Los escenarios de la globalizacion y la pobreza en Mexico.
Rabenoro, T. (France): Leveraging experts’ knowledge to improve diagnostics on aircraft
engine.
Reyes, H. (Mexico): Cálculos de los tamaños de prueba y niveles de significancia ajustados
para la prueba de Blackwelder
Reyes-Cervantes, H. (Mexico): Modelación estadística sobre el egreso de universitarios de
ciencias exactas de la FCFM-BUAP.
Rueda, M del M. (Spain): Un estudio estadístico sobre el uso problemático del móvil y
personalidad.
Rynkiewicz, J. (France): Assessment of the influence of education level on voting intention
for the extreme right in France.
Salgado, L. (Mexico): Evaluación de la pobreza, desarrollo comunitario o desarrollo local y
el municipio.
Sánchez, A. (Mexico): Employment of new graduates in the Mexican labor market (Case
Study, University of the State of Mexico).
Sánchez, A. (Mexico): Econometric & statistical response model applied to socioeconomical careers at UAEM. C. U. Valle de México.
Santiago, A. (Mexico): Ratio type estimation using the knowledge of the auxiliary variable
for ranking and estimating.
Sarin, R. K. (USA): Guided decision processes.
Sautto, J.M. (Mexico): Ratio type estimation using the knowledge of the auxiliary variable
for ranking and estimating.
Sistachs, V. (Cuba): Promedio Bayesiano de Modelos en Salud: una revisión.
Skala, V. (Czech Republic): A practical use of radial basis functions interpolation and
approximation.
Sombra Medeiros, F. (Brazil): Processing: applications in health care and remote sensing
of environment.
Tamayo Garcia, A. (Cuba): Diseño de un procedimiento para la planificación y control de
la producción haciendo uso de herramientas matemáticas.
Tuero A. (Cuba): Meta-analysis of safety and efficacy in patients with diabetic foot ulcers.
International projection of Heberprot-p®.
Uranga, R. (Cuba): Simulación para evaluar el tratamiento de datos faltantes en modelos
de transición en el contexto de ensayos clínicos.
Ursprung, H. W. (Germany): Participant.
Vázquez Alfonso, Y. (Cuba): Hiperentorno basado en objetos de aprendizaje destinado a
la asignatura Análisis de Datos (Estadística) para la Carrera de Cultura Física.
Viada, C. (Cuba): Tiempo al deterioro de los scores de calidad de vida como una
modalidad de análisis longitudinal para pacientes con cáncer de cabeza y cuello tratados
con Nimotuzumab.
Wickström, B. A (Hungary): Language rights: a welfare-economics approach.
Zabala Corrales, J. J. (Colombia): Population sampling fuzzy.
Notes
Notes
Notes
Notes
Descargar