<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:dc="http://purl.org/dc/elements/1.1/" version="2.0">
<channel>
<title>Volumen 20 | Número 01</title>
<link>http://sedici.unlp.edu.ar:80/handle/10915/97185</link>
<description/>
<pubDate>Wed, 18 Nov 2026 19:09:10 GMT</pubDate>
<dc:date>2026-11-18T19:09:10Z</dc:date>
<item>
<title>Invariance and Same-Equivariance Measures for Convolutional Neural Networks</title>
<link>http://sedici.unlp.edu.ar:80/handle/10915/97204</link>
<description>Revision
Journal of Computer Science &amp; Technology; vol. 20, no. 1; https://sedici.unlp.edu.ar/handle/10915/90903
Our main objective in this thesis is to contribute to the understanding and improvement of equivariance in neural&#13;
network models. In terms of applications, we focus on handshape classification for sign language and other types&#13;
of gestures using convolutional networks. Therefore, we set the following specific goals:&#13;
• Analyze CNN models design specifically for equivariance&#13;
• Compare specific models and data augmentation as means to obtain equivariance. Evaluate transfer learning strategies to obtain equivariant models starting with non-equivariant ones.&#13;
• Develop equivariance measures for activations or inner representations in Neural Networks. Implement those measures in an open source library. Analyze the measures behavior, and compare with existing measures.
</description>
<pubDate>Fri, 01 May 2020 00:00:00 GMT</pubDate>
<guid isPermaLink="false">http://sedici.unlp.edu.ar:80/handle/10915/97204</guid>
<dc:date>2020-05-01T00:00:00Z</dc:date>
<dc:description>Our main objective in this thesis is to contribute to the understanding and improvement of equivariance in neural&#13;
network models. In terms of applications, we focus on handshape classification for sign language and other types&#13;
of gestures using convolutional networks. Therefore, we set the following specific goals:&#13;
• Analyze CNN models design specifically for equivariance&#13;
• Compare specific models and data augmentation as means to obtain equivariance. Evaluate transfer learning strategies to obtain equivariant models starting with non-equivariant ones.&#13;
• Develop equivariance measures for activations or inner representations in Neural Networks. Implement those measures in an open source library. Analyze the measures behavior, and compare with existing measures.</dc:description>
</item>
<item>
<title>Political Alignment Identification: a Study with Documents of Argentinian Journalists</title>
<link>http://sedici.unlp.edu.ar:80/handle/10915/97203</link>
<description>Articulo
Identificación de alineamiento político: un estudio con documentos de periodistas argentinos
Journal of Computer Science &amp; Technology; vol. 20, no. 1
Political alignment identification is an author profiling task that aims at identifying political bias/orientation in people’ writings. As usual in any automatic text analysis, a critical aspect here is having available adequate data sets so that the data mining and machine learning approaches can obtain reliable and informative results.&#13;
This article makes a contribution in this regard by presenting a new corpus for the study of political alignment in documents of Argentinian journalists. The study also includes several kinds of analysis of documents of pro-government and opposition journalists such as the relevance of terms in each journalist class, sentiment analysis, topic modelling and the analysis of psycholinguistic indicators obtained from the Linguistic Inquiry and Word Count (LIWC) system. From the experimental results, interesting patterns could be observed such as the topics both types of journalists write about, how the sentiment polarities are distributed and how the writings of pro-government and opposition journalists differ in the distinct LIWC categories.
</description>
<pubDate>Fri, 01 May 2020 00:00:00 GMT</pubDate>
<guid isPermaLink="false">http://sedici.unlp.edu.ar:80/handle/10915/97203</guid>
<dc:date>2020-05-01T00:00:00Z</dc:date>
<dc:description>Political alignment identification is an author profiling task that aims at identifying political bias/orientation in people’ writings. As usual in any automatic text analysis, a critical aspect here is having available adequate data sets so that the data mining and machine learning approaches can obtain reliable and informative results.&#13;
This article makes a contribution in this regard by presenting a new corpus for the study of political alignment in documents of Argentinian journalists. The study also includes several kinds of analysis of documents of pro-government and opposition journalists such as the relevance of terms in each journalist class, sentiment analysis, topic modelling and the analysis of psycholinguistic indicators obtained from the Linguistic Inquiry and Word Count (LIWC) system. From the experimental results, interesting patterns could be observed such as the topics both types of journalists write about, how the sentiment polarities are distributed and how the writings of pro-government and opposition journalists differ in the distinct LIWC categories.</dc:description>
</item>
<item>
<title>Parallelism and Hybridization in Differential Evolution to solve the Flexible Job Shop Scheduling Problem</title>
<link>http://sedici.unlp.edu.ar:80/handle/10915/97202</link>
<description>Articulo
Paralelismo e hibridización en un algoritmo de evolución diferencial para resolver el problema de planificación job shop flexible
Journal of Computer Science &amp; Technology; vol. 20, no. 1
The Flexible Job Shop Scheduling Problem (FJSSP) is one of the most challenging combinatorial optimization problems, with practical applicability in a real production environment. In this work, we propose a simple Differential Evolution (DE) algorithm to tackle this problem. To represent a FJSSP solution, a real value representation is adopted, which requires a very simple conversion mechanism to obtain a feasible schedule. Consequently, the DE algorithm still works on the continuous domain to explore the problem search space of the discrete FJSSP. Moreover, to enhance the local search ability and to balance the exploration and exploitation capabilities, a simple local search algorithm is embedded in the DE framework.&#13;
Also, the parallelism of the DE operations is included to improve the efficiency of the whole algorithm. Experimental results confirm the significant improvement achieved by integrating the modifications introduced in this study. Additionally, test results show that our algorithm is competitive when compared with most existing approaches for FJSSP.
</description>
<pubDate>Fri, 01 May 2020 00:00:00 GMT</pubDate>
<guid isPermaLink="false">http://sedici.unlp.edu.ar:80/handle/10915/97202</guid>
<dc:date>2020-05-01T00:00:00Z</dc:date>
<dc:description>The Flexible Job Shop Scheduling Problem (FJSSP) is one of the most challenging combinatorial optimization problems, with practical applicability in a real production environment. In this work, we propose a simple Differential Evolution (DE) algorithm to tackle this problem. To represent a FJSSP solution, a real value representation is adopted, which requires a very simple conversion mechanism to obtain a feasible schedule. Consequently, the DE algorithm still works on the continuous domain to explore the problem search space of the discrete FJSSP. Moreover, to enhance the local search ability and to balance the exploration and exploitation capabilities, a simple local search algorithm is embedded in the DE framework.&#13;
Also, the parallelism of the DE operations is included to improve the efficiency of the whole algorithm. Experimental results confirm the significant improvement achieved by integrating the modifications introduced in this study. Additionally, test results show that our algorithm is competitive when compared with most existing approaches for FJSSP.</dc:description>
</item>
<item>
<title>Tuning a hybrid SA based algorithm applied to Optimal Sensor Network Design</title>
<link>http://sedici.unlp.edu.ar:80/handle/10915/97201</link>
<description>Articulo
Ajustes de un algoritmo híbrido basado en SA aplicado al diseño óptimo de redes de sensores
Journal of Computer Science &amp; Technology; vol. 20, no. 1
Sensor network design problem (SNDP) in process plants includes the determination of which process variables should be measured to achieve a required degree of knowledge about the plant. We propose to solve the SNDP problem in plants of increasing size and complexity using a hybrid algorithm based on Simulated Annealing (HSA) as main metaheuristic and Tabu Search embedded with Strategic Oscillation (SOTS) as a subordinate metaheuristic. We studied the tuning of control parameters in order to improve the HSA performance. Experimental results indicate that a high-quality solution in reasonable computational times can be found by HSA effectively. Moreover, HSA shows good features solving SNDP compared with proposals from the literature.
</description>
<pubDate>Fri, 01 May 2020 00:00:00 GMT</pubDate>
<guid isPermaLink="false">http://sedici.unlp.edu.ar:80/handle/10915/97201</guid>
<dc:date>2020-05-01T00:00:00Z</dc:date>
<dc:description>Sensor network design problem (SNDP) in process plants includes the determination of which process variables should be measured to achieve a required degree of knowledge about the plant. We propose to solve the SNDP problem in plants of increasing size and complexity using a hybrid algorithm based on Simulated Annealing (HSA) as main metaheuristic and Tabu Search embedded with Strategic Oscillation (SOTS) as a subordinate metaheuristic. We studied the tuning of control parameters in order to improve the HSA performance. Experimental results indicate that a high-quality solution in reasonable computational times can be found by HSA effectively. Moreover, HSA shows good features solving SNDP compared with proposals from the literature.</dc:description>
</item>
<item>
<title>Green High Performance Simulation for AMB models of Aedes aegypti</title>
<link>http://sedici.unlp.edu.ar:80/handle/10915/97193</link>
<description>Articulo
Simulación green de alto rendimiento de un modelo basado en agentes del mosquito Aedes aegypti
Journal of Computer Science &amp; Technology; vol. 20, no. 1
The increase in temperature caused by the climate change has resulted in the rapid dissemination of infectious diseases. Given the alert for the current situation, the World Health Organization (WHO) has declared a state of health emergency, highlighting the severity of the situation in some countries. For this reason, coming up with knowledge and tools that can help control and eradicate the vectors propagating these diseases is of the utmost importance. Highperformance modeling and simulation can be used to produce knowledge and strategies that allow predicting infections, guiding actions and/or training health/civil protection agents. The model developed as part of this research work is aimed at assisting the decision-making process for disease prevention and control, as well as evaluating the reproduction and predicting the evolution of the Aedes aegypti mosquito, which is the transmitting vector of the dengue, Zika and chikungunya diseases. Decisionmaking based on these models requires a large number of simulations to achieve results with statistical variability.&#13;
The objective of this paper is to demonstrate that the GPU is a suitable platform from the point of view of the reduction of energy consumed for HPC simulations. It is also shown that it is possible to define energy prediction models that allow scientists to plan their experiments based on energy consumption and select those that are representative for decision making by reducing energy consumption in HPC simulations.; El aumento de la temperatura a raíz del cambio climático, ha dado lugar a la rápida expansión de enfermedades infecciosas. Dada la alerta por la situación actual, la Organización Mundial de la Salud (OMS) ha declarado la emergencia sanitaria poniendo de manifiesto la grave situación que se vive en algunos países. Es por ello que es necesario aportar conocimiento y herramientas que ayuden al control y erradicación del vector que propaga estas enfermedades. El modelado y la simulación de altas prestaciones pueden ayudar a aportar conocimiento y estrategias que permitan predecir infecciones, orientar actuaciones y/o formar a los agentes de protección civil/salud. El modelo desarrollado en este trabajo, tiene por objetivo ayudar a la toma de decisiones de prevención y control, a evaluar la reproducción y a predecir la evolución del mosquito Aedes aegypti, transmisor de las enfermedades dengue, Zika y chikungunya. Dado que son necesarias un elevado número de simulaciones para tener resultados con variabilidad estadística, se ha utilizado GPU. Con esta plataforma se busca: su potencia de cómputo para reducir el tiempo de ejecución y, además, reducir el consumo de energía.&#13;
Para ello se proponen diferentes escenarios y experimentos para comprobar los beneficios de la arquitectura propuesta.
</description>
<pubDate>Fri, 01 May 2020 00:00:00 GMT</pubDate>
<guid isPermaLink="false">http://sedici.unlp.edu.ar:80/handle/10915/97193</guid>
<dc:date>2020-05-01T00:00:00Z</dc:date>
<dc:description>The increase in temperature caused by the climate change has resulted in the rapid dissemination of infectious diseases. Given the alert for the current situation, the World Health Organization (WHO) has declared a state of health emergency, highlighting the severity of the situation in some countries. For this reason, coming up with knowledge and tools that can help control and eradicate the vectors propagating these diseases is of the utmost importance. Highperformance modeling and simulation can be used to produce knowledge and strategies that allow predicting infections, guiding actions and/or training health/civil protection agents. The model developed as part of this research work is aimed at assisting the decision-making process for disease prevention and control, as well as evaluating the reproduction and predicting the evolution of the Aedes aegypti mosquito, which is the transmitting vector of the dengue, Zika and chikungunya diseases. Decisionmaking based on these models requires a large number of simulations to achieve results with statistical variability.&#13;
The objective of this paper is to demonstrate that the GPU is a suitable platform from the point of view of the reduction of energy consumed for HPC simulations. It is also shown that it is possible to define energy prediction models that allow scientists to plan their experiments based on energy consumption and select those that are representative for decision making by reducing energy consumption in HPC simulations.

El aumento de la temperatura a raíz del cambio climático, ha dado lugar a la rápida expansión de enfermedades infecciosas. Dada la alerta por la situación actual, la Organización Mundial de la Salud (OMS) ha declarado la emergencia sanitaria poniendo de manifiesto la grave situación que se vive en algunos países. Es por ello que es necesario aportar conocimiento y herramientas que ayuden al control y erradicación del vector que propaga estas enfermedades. El modelado y la simulación de altas prestaciones pueden ayudar a aportar conocimiento y estrategias que permitan predecir infecciones, orientar actuaciones y/o formar a los agentes de protección civil/salud. El modelo desarrollado en este trabajo, tiene por objetivo ayudar a la toma de decisiones de prevención y control, a evaluar la reproducción y a predecir la evolución del mosquito Aedes aegypti, transmisor de las enfermedades dengue, Zika y chikungunya. Dado que son necesarias un elevado número de simulaciones para tener resultados con variabilidad estadística, se ha utilizado GPU. Con esta plataforma se busca: su potencia de cómputo para reducir el tiempo de ejecución y, además, reducir el consumo de energía.&#13;
Para ello se proponen diferentes escenarios y experimentos para comprobar los beneficios de la arquitectura propuesta.</dc:description>
</item>
<item>
<title>A T-cell algorithm for solving dynamic economic power dispatch problems</title>
<link>http://sedici.unlp.edu.ar:80/handle/10915/97190</link>
<description>Articulo
Un algoritmo basado en células T para resolver problemas de despacho de energía económico y dinámico
Journal of Computer Science &amp; Technology; vol. 20, no. 1
This paper presents the artificial immune system IA DED (Immune Algorithm Dynamic Economic Dispatch) to solve the Dynamic Economic Dispatch (DED) problem and the Dynamic Economic Emission Dispatch (DEED) problem. Our approach considers these as dynamic problems whose constraints change over time. IA DED is inspired on the activation process that T cells suffer in order to find partial solutions. The proposed approach is validated using several DED problems taken from specialized literature and one DEED problem. The latter is addressed by transforming a multi-objective problem into a single-objective problem by using a linear aggregating function that combines the (weighted) values of the objectives into a single scalar value. Our results are compared with respect to those obtained by other approaches taken from the specialized literature.&#13;
We also provide some statistical analysis in order to determine the sensitivity of the performance of our proposed approach to its parameters. Part of this work was presented at the XXV Argentine Congress of Computer Science (CACIC), 2019.
</description>
<pubDate>Fri, 01 May 2020 00:00:00 GMT</pubDate>
<guid isPermaLink="false">http://sedici.unlp.edu.ar:80/handle/10915/97190</guid>
<dc:date>2020-05-01T00:00:00Z</dc:date>
<dc:description>This paper presents the artificial immune system IA DED (Immune Algorithm Dynamic Economic Dispatch) to solve the Dynamic Economic Dispatch (DED) problem and the Dynamic Economic Emission Dispatch (DEED) problem. Our approach considers these as dynamic problems whose constraints change over time. IA DED is inspired on the activation process that T cells suffer in order to find partial solutions. The proposed approach is validated using several DED problems taken from specialized literature and one DEED problem. The latter is addressed by transforming a multi-objective problem into a single-objective problem by using a linear aggregating function that combines the (weighted) values of the objectives into a single scalar value. Our results are compared with respect to those obtained by other approaches taken from the specialized literature.&#13;
We also provide some statistical analysis in order to determine the sensitivity of the performance of our proposed approach to its parameters. Part of this work was presented at the XXV Argentine Congress of Computer Science (CACIC), 2019.</dc:description>
</item>
</channel>
</rss>
