Universidad del Norte, Colombia

About: Universidad del Norte, Colombia is a education organization based out in Barranquilla, Colombia. It is known for research contribution in the topics: Population & Context (language use). The organization has 3562 authors who have published 4355 publications receiving 37861 citations. The organization is also known as: University of the North, Colombia & Uninorte.

Support vector machine in statistical process monitoring: a methodological and analytical review

Abstract: In statistical process monitoring, data mining algorithms are applied for control chart pattern recognition (CCPR) not only to detect but also to identify abnormal patterns associated with assignable causes. Recently, the use of support vector machines (SVMs) has achieved remarkable results in statistical process control applications due to its excellent generalization performance. Although there is a lot of research that highlights the superiority of support vector machine over other algorithms used in systems based on data mining control, there are no studies that analyze the design elements of a SVM classifier. Consequently, a comprehensive review, classification and an analysis of information found in the literature are presented in this paper regarding to the SVM classifier. The aim of this research is to provide researchers a starting point to potentiate the performance of the SVM classifier for assuring the best possible classification and improving the detection efficiency. Sixty-one research articles from 2001 to 2015 are critically analyzed based on their methodology following a classification scheme derived from the support vector classification framework. The analysis showed that the feature extraction and selection play a crucial role on the performance of classifiers. Studies revealed that using extracted features or a combination of raw data and feature extraction achieves better classification performances than using only raw data as an input. On the other hand, there is an ample gap to extend the research on internal structure configurations under various types of kernels, multi-classification, and ensemble approaches. Finally, this paper evidences that the application of nature inspired algorithms for kernel parameter selection in auto-correlated SVM-based process monitoring systems remains unexplored.

Efficient extraction of human motion volumes by tracking

Abstract: We present an automatic and efficient method to extract spatio-temporal human volumes from video, which combines top-down model-based and bottom-up appearance-based approaches. From the top-down perspective, our algorithm applies shape priors probabilistically to candidate image regions obtained by pedestrian detection, and provides accurate estimates of the human body areas which serve as important constraints for bottom-up processing. Temporal propagation of the identified region is performed with bottom-up cues in an efficient level-set framework, which takes advantage of the sparse top-down information that is available. Our formulation also optimizes the extracted human volume across frames through belief propagation and provides temporally coherent human regions. We demonstrate the ability of our method to extract human body regions efficiently and automatically from a large, challenging dataset collected from YouTube.

Assessing Vulnerability of Transportation Networks for Disaster Response Operations

Abstract: Transportation networks and systems are vulnerable to natural disasters. During disaster response operations, the degraded functionality of the system can negatively impact the affected population because disrupting relief activitiesincreases human suffering resulting from the lack of access to essential goods or services. Mathematical formulations for assessing transportation network vulnerability do not generally consider this lack of access or deprivation costs, and can lead to inappropriate strategies for humanitarian assistance. This paper proposes a transportation network vulnerability assessment model that allows identifying critical links for the development of high impact disaster response operations. The model is based on an economic analysis that considers the logistical costs of the distribution operations and the external effects derived from the delays in the provision of basic supplies (deprivation costs). The approach is particularly useful for planning resilient disaster response plans in the preparednessstage, prioritizing investment for mitigation and adaptation, and prioritizing the rehabilitation (access restoration) of the disrupted links in the response and recovery stages. In addition to numerical experiments using case study networks, the authors implemented the model to the coffee-producing region of Colombia, which was hit by an earthquake in 1999.

Women's perceived risk of sexual harassment in a Bus Rapid Transit (BRT) system: The case of Barranquilla, Colombia

Abstract: Background : Sexual harassment in public transportation is a growing concern, particularly among women. Over the years, there have been several programs and policies to mitigate sexual harassment while using public transport. However, there is little evidence of the effectiveness of these strategies, especially in Latin America. Objective : This investigation aims to determine the factors that influence women's perceived risk of sexual harassment while using public transport in Colombia. Methods : In this study, we designed an image-based stated preferences survey based on the current bus rapid transit (BRT) system in Barranquilla, Colombia. Several variables were con- sidered in this experiment including the time of the day, surveillance, and crowding, among others. For each scenario, participants reported whether they felt safe or not. Then, a logistic regression analysis was conducted to identify the factors that influence women's perceived risk of sexual harassment while using the BRT system. Results : The results show that more than 60% of respondents have been a victim of sexual harassment while using the BRT system. Also, overcrowded buses proved to have the most negative effect on the perceived risk of sexual harassment. Travelling at night, lighting and being alone were all significant variables as well. Conclusion : Sexual harassment could potentially influence use of the BRT. The findings of this research can be used to develop countermeasures and increase public transport ridership.

High-Throughput Sequencing Reveals Circulating miRNAs as Potential Biomarkers of Kidney Damage in Patients with Systemic Lupus Erythematosus.

Abstract: Renal involvement is one of the most severe manifestations of systemic lupus erythematosus (SLE). Renal biopsy is the gold standard when it comes to knowing whether a patient has lupus nephritis, and the degree of renal disease present. However, the biopsy has various complications, bleeding being the most common. Therefore, the development of alternative, non-invasive diagnostic tests for kidney disease in patients with SLE is a priority. Micro RNAs (miRNAs) are differentially expressed in various tissues, and changes in their expression have been associated with several pathological processes. The aim of this study was to identify changes in the abundance of miRNAs in plasma samples from patients with lupus nephritis that could potentially allow the diagnosis of renal damage in SLE patients. This is an observational case-control cross-sectional study, in which we characterized the differential abundance profiles of miRNAs among patients with different degrees of lupus compared with SLE patients without renal involvement and healthy control individuals. We found 89 miRNAs with changes in their abundance between lupus nephritis patients and healthy controls, and 17 miRNAs that showed significant variations between SLE patients with or without renal involvement. Validation for qPCR of a group of miRNAs on additional samples from lupus patients with or without nephritis, and from healthy individuals, showed that five miRNAs presented an average detection sensitivity of 97%, a specificity of 70.3%, a positive predictive value of 82.5%, a negative predictive value of 96% and a diagnosis efficiency of 87.9%. These results strongly suggest that miR-221-5p, miR-380-3p, miR-556-5p, miR-758-3p and miR-3074-3p are potential diagnostic biomarkers of lupus nephritis in patients with SLE. The observed differential pattern of miRNA abundance may have functional implications in the pathophysiology of SLE renal damage.