Rivas-Perea P.,Baylor University |
Cota-Ruiz J.,Autonomous University of Ciudad Juarez
Pattern Recognition Letters | Year: 2013
This paper presents a method to train a Support Vector Regression (SVR) model for the large-scale case where the number of training samples supersedes the computational resources. The proposed scheme consists of posing the SVR problem entirely as a Linear Programming (LP) problem and on the development of a sequential optimization method based on variables decomposition, constraints decomposition, and the use of primal-dual interior point methods. Experimental results demonstrate that the proposed approach has comparable performance with other SV-based classifiers. Particularly, experiments demonstrate that as the problem size increases, the sparser the solution becomes, and more computational efficiency can be gained in comparison with other methods. This demonstrates that the proposed learning scheme and the LP-SVR model are robust and efficient when compared with other methodologies for large-scale problems. © 2012 Elsevier B.V. All rights reserved.
Vital C.,Autonomous University of Ciudad Juarez |
Martins E.P.,Indiana University Bloomington
PLoS ONE | Year: 2013
Certain individuals are more effective than others at using individual experience to impact group behavior. Here, we tested whether pre-training of zebrafish that are at the focal central of social group dynamics ("Key" fish) has a stronger positive impact on group performance than does pre-training of less central ("Non-Key") fish. We used very short observation periods and social network statistics to identify Key and Non-Key individuals, trained these fish to respond to an aversive stimulus, and then measured group performance after returning these now-experienced fish to a social setting. Although Key and Non-Key fish evaded the stimulus equally quickly as individuals, groups with experienced Key fish escaped the aversive stimulus more quickly than did groups with experienced Non-Key fish. The impact depended on genetic background: PN zebrafish on the social extremes (more often males) influenced the group's baseline response to the aversive stimulus, whereas experienced Scientific Hatcheries' zebrafish (both males and females) influenced the change in response over repeated trials. These results suggest that social roles are an important feature of information transfer across a group, and set the stage for future research into the genetic and evolutionary basis of social learning. © 2013 Vital and Martins.
Hernandez V.H.,Autonomous University of Ciudad Juarez
Revista Panamericana de Salud Publica/Pan American Journal of Public Health | Year: 2012
Objective. Prepare a tool for the exploratory study of road accidents in Ciudad Juarez, Chihuahua, Mexico, that exclusively applies the spatial geographical variable (location). Methods. Observational and cross-sectional study that uses a Geographic Information System to explore the spatial nature of 13 305 road accidents recorded during 2008 and 2009 in Ciudad Juarez. Indicators were constructed that approximated the transit flow and included two variables: indices of the level of urbanization and population density. Results. The value of the global spatial autocorrelation was positive, indicating the presence of groupings that were identified through the spatial association indicators. There are road risk clusters located in areas with a high level of urbanization, low population density, and a high transit flow level. Conclusions. The exploratory analysis of spatial data is a phase that precedes the use of multivariate techniques with a broader scope. The application of exploratory analysis techniques in itself makes it possible to standardize spatial groupings, identify global autocorrelation, and indicate the direction of the variables under study.
Rodriguez-Picon L.A.,Autonomous University of Ciudad Juarez
International Journal of Advanced Manufacturing Technology | Year: 2016
Any organization can be defined as a system whose control is governed by the flow of information across all systems to which it belongs. Information is defined as an infinite-dimensional entity that is constantly moving and being assessed through mechanisms and control systems in order to achieve a goal. This article presents an approach for design of parameters based on information content. The development of a product is evaluated from the initial concept to the optimal design of its production parameters. In this approach, the values of a set of design parameters related to functional requirements are considered. The behavior of these functional requirements is modeled by a Gaussian distribution, which represents the design distribution. Once the product is produced in mass, a Gaussian distribution also models the behavior of the functional requirements and represents the process distribution. The intersection between distributions can be seen as a common range that describes the compliance with the functional requirements. As the process distribution approaches the most to the design distribution, the best functional requirements are met. Best levels of the parameters of the case study were defined based on this approach, which enables to reduce the variation on the response variable. © 2016 Springer-Verlag London
Diaz-Caravantes R.E.,Autonomous University of Ciudad Juarez
Geographical Journal | Year: 2012
The peri-urban area is the region where there is a more dynamic interaction between the urban and rural. The peri-urban area supplies natural resources, such as land for urban expansion and agricultural products to feed the urban population. In arid and semi-arid lands, such as northern Mexico, these areas may also be the source of water for the city's domestic demand. In addition, scholars argue that peri-urban residents may have a more advantageous geographical position for selling their labour and agricultural products in cities and, by doing so, sustaining their livelihoods. A considerable number of studies have examined the peri-urban to urban natural resources transfer in terms of land annexation, housing construction, and infrastructure issues; however, the study of the effects of the reallocation of peri-urban water resources to serve urban needs is critical as well because the livelihoods of peri-urban residents, such as those based on agriculture and livestock, depend on water availability. In the case of Hermosillo there is a tremendous pressure on the water resources of peri-urban small farm communities or ejidos because of urban demand. Based on interviews and structured surveys with producers and water managers, this paper examines how peri-urban livelihoods have been reshaped by the reallocation of the city's natural resources in many cases caused some ejido members or ejidatarios to lose livelihoods. © 2011 The Author. The Geographical Journal © 2011 Royal Geographical Society (with the Institute of British Geographers).