Juan Velasquez - Academia.edu (original) (raw)
Papers by Juan Velasquez
Springer Handbook of Automation, 2009
ABSTRACT This chapter contains automation information collected from many different sources aroun... more ABSTRACT This chapter contains automation information collected from many different sources around the world, among them the US Census Bureau, the International Federation of Automatic Control (IFAC), the American Automatic Control Council (AACC), the American Bankers Association, World Robotics, and many more. Section 1 introduces automation statistical data according to its application domain, such as: financial and e-commerce, industrial, healthcare, and the service industries. In Section 2 a list of worldwide automation, control, and robotics associations is provided. Section 3 provides the reader with a list of automation labs around the world and lastly, in Section 4 a list of automation related journals and publications is presented. The statistics, organizations, and journals included in this chapter open a window into the current and emerging state of automation. While attempting to be comprehensive, because of the broadness of automation, the authors recognize that there may be additional relevant items, but none were omitted intentionally. The scope of the information is meant to highlight and expose the broad span, applications, concerns and benefits of automation, and clearly, cannot completely include all areas of automation influence. Beyond this chapter, other chapters in this handbook provide additional statistical data directly related to their specific topic. For the complete review see Zbl 1218.93004.
Studies in Computational Intelligence, 2013
End users leave traces of behavior all over the Web all times. From the explicit or implicit feed... more End users leave traces of behavior all over the Web all times. From the explicit or implicit feedback of a multimedia document or a comment in an online social network, to a simple click in a relevant link in a search engine result, the information that we as users pour into the Web defines its actual representation, which is independent for each user. Our usage can be represented by different sources of data, for which different collection strategies must be considered, as well as the merging and cleaning techniques for Web usage data. Once the data is properly preprocessed, the identification of an individual user within the Web can be a complex task. Understanding the whole life of a user within a session in a Web site and the path that was pursued involves advanced data modeling and a set of assumptions which are modified every day, as new ways to interact with the online content are created. The objective is to understand the behaviour and preferences of a web user, also when several privacy issues are involved, which, as of today, are not clear how to be properly addressed. In this chapter, all previous topics regarding the processing of Web usage data are extensively discussed.
Proceedings of the first international conference on Autonomous agents - AGENTS '97, 1997
Google, Inc. (search), Subscribe (Full Service), Register (Limited Service, Free), Login. Search:... more Google, Inc. (search), Subscribe (Full Service), Register (Limited Service, Free), Login. Search: The ACM Digital Library The Guide. ...
2012 7th Colombian Computing Congress (CCC), 2012
ABSTRACT In this article, we evaluate the autoregressive neural networks (ARNN) in forecasting ti... more ABSTRACT In this article, we evaluate the autoregressive neural networks (ARNN) in forecasting time series with trend and seasonal cycle. For the time series analyzed, we found that the combined application of simple and seasonal differentiation factors does not necessarily contribute to better forecasts than if these are applied separately, and that while the preprocessing of data helps to better forecast does not mean that models are not able to capture both the trend and the seasonal cycle present in the untransformed series. In this regard, both the ARNN as the MLP (used as a benchmark) gave better results than the SARIMA process, which in the case of MLP contradicts what is stated in [15], so the assertion that MLPs do not capture these components applies only in special cases. This conclusion can be extended to ARNN model.
IEEE Latin America Transactions, 2014
2014 IEEE Congress on Evolutionary Computation (CEC), 2014
Lecture Notes in Computer Science, 2014
2010 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology, 2010
A biologically inspired cognitive model is presented for human decision making and applied to the... more A biologically inspired cognitive model is presented for human decision making and applied to the simulation of the web user. The model is based on the Neurophysiology description of multiple decision process; this is a well proven psychological theory. The model simulates the behaviour of a real user on a website and it was observed that the distribution of artificial web users in sessions successfully simulates a genuine user's web mode of behaviour. On the hypothesis that the adjusted artificial web user behaves statistically similar to the human web users, a system was created for the improvement of the structure of a web site based on stochastic simulations as a Proof of Concept. Since simulation recover observed statistical behaviour, changes on a web site are used to predict changes on navigational patterns.
2012 7th Colombian Computing Congress (CCC), 2012
ABSTRACT In the practice, to find the parameters for nonlinear Moving average(NLMA) models is dif... more ABSTRACT In the practice, to find the parameters for nonlinear Moving average(NLMA) models is difficult because there are not analytical way to solve maximum likelihood function. In this paper, we compare the parameters estimated for NLMA model with Supernova and PSO-DE algorithm. We found that Supernova estimation is better in terms of a reduction of error and nearest solutions to optimal point.
DYNA, 2014
The Committee on Publication Ethics (COPE) -conformed by the main scientific publishers-has warne... more The Committee on Publication Ethics (COPE) -conformed by the main scientific publishers-has warned about the increasing number of ethical problems in scientific publication and recent scandals seem to indicate that ethical misconduct is repetitive. Ethical problems in scientific publication arise when the person deviates from expected moral behavior. Misconduct may be explained, at least in part, because many postgraduate students and young researchers seem to understand that ethical problems are only related to plagiarism of complete works or duplication of publications, and because it seems to be a lack of knowledge of the ethical standards in scientific publication. However, there are many other aspects conducing to ethical problems. The objective of this paper is to discuss and spread the ethical position of the main scientific publishers and researchers with the aim of build a unified point of view. In this paper, seventeen tips for avoiding ethical problems in scientific publication are presented, explained and discussed. I hope that this work will be valuable for postgraduate students and young researchers and answers many common questions about ethics in scientific publication.
The Information Society, 2010
Google, Inc. (search), Subscribe (Full Service), Register (Limited Service, Free), Login. Search:... more Google, Inc. (search), Subscribe (Full Service), Register (Limited Service, Free), Login. Search: The ACM Digital Library The Guide. ...
International Journal of Production Economics, 2008
In this work, a new method to resolve conflict situations (Mcr) common in collaborative facility ... more In this work, a new method to resolve conflict situations (Mcr) common in collaborative facility design using computer support tools is developed. The method uses traditional human conflict resolution approaches that have been used successfully by others. It incorporates ...
International Journal of Computer Integrated Manufacturing, 2008
In collaborative e-work over largely distributed organisations it is often necessary to match set... more In collaborative e-work over largely distributed organisations it is often necessary to match sets of components or suppliers for better-quality assemblies, or for better-quality contracts. In such cases, the importance of efficiently and effectively sharing, matching and leveraging real-time information is demonstrated in e-work and e-manufacturing systems via a best-matching protocol (BMP). The developed BMP enables better and quicker matches for both geometrical (e.g. physical parts) and non-geometrical attributes (e.g. members in a supply network). Improvement is measured using the newly defined best-fit index. Four statistical distributions are applied to characterise part specifications of the physical components. The statistical analysis corroborates that the performance of BMP in yielding significantly better matches for geometrical specifications (within acceptable tolerance limits) is consistent across uniform and normal distributions. The economic value of a match (EVM) is also evaluated for different economic costs. The results of applying EVM indicate that organisations need to define policies for matching as it is not always economically desirable to match parts. A double-match that includes a best-matching process for non-geometrical attributes and prior matching results from the manufacturing scenario is also performed to identify suppliers that provide the best parts and costs combinations.
Ingeniare. Revista chilena de ingeniería, 2010
La predicción de series de tiempo es un importante problema de investigación debido a sus implica... more La predicción de series de tiempo es un importante problema de investigación debido a sus implicaciones en ingeniería, economía, finanzas y ciencias sociales. Un importante tópico de esta problemática es el desarrollo de nuevos modelos y su comparación con aproximaciones previas en términos de la precisión del pronóstico. Recientemente, las máquinas de vectores de soporte (SVM) han sido usadas para la predicción de series de tiempo, pero las experiencias reportadas son limitadas y hay algunos problemas relacionados con su especificación. El objetivo de este artículo es proponer una técnica novedosa para estimar algunas constantes en las SVM que usualmente son fijadas en forma empírica por el modelador. La técnica propuesta es usada para estimar varias SVM con el fin de pronosticar cinco series benchmark; los resultados obtenidos son comparados con las estadísticas reportadas en otros artículos. La metodología propuesta permite obtener SVM competitivas para las series pronosticadas en comparación con los resultados obtenidos usando otros modelos más tradicionales.
Ingeniare. Revista chilena de ingeniería, 2007
En este trabajo se propone la representación de la dinámica del Índice de Oscilación del Sur usan... more En este trabajo se propone la representación de la dinámica del Índice de Oscilación del Sur usando una nueva clase de modelo híbrido no lineal. En este nuevo modelo, la no-linealidad en la media es representada usando un sistema adaptativo neurodifuso de inferencia (ANFIS), mientras que la varianza es representada usando una componente autoregresiva heterocedástica condicional. En adición, una estrategia de especificación es desarrollada para el modelo propuesto, la cual está basada en una batería de pruebas estadísticas. Los resultados indican que el modelo propuesto captura de mejor forma las principales características de la serie de tiempo estudiada, en comparación con otros modelos competitivos.
Ingeniare. Revista chilena de ingeniería, 2008
Una red neuronal autorregresiva es estimada para el precio mensual brasileño de corto plazo de la... more Una red neuronal autorregresiva es estimada para el precio mensual brasileño de corto plazo de la electricidad, la cual describe mejor la dinámica de los precios que un modelo lineal autorregresivo y que un perceptrón multicapa clásico que usan las mismas entradas y neuronas en la capa oculta. El modelo propuesto es especificado usando un procedimiento estadístico basado en el contraste del radio de verosimilitud. El modelo pasa una batería de pruebas de diagnóstico. El procedimiento de especificación propuesto permite seleccionar el número de unidades en la capa oculta y las entradas a la red neuronal, usando pruebas estadísticas que tienen en cuenta la cantidad de los datos y el ajuste del modelo a la serie de precios. La especificación del modelo final demuestra que el precio para el próximo mes es una función no lineal del precio actual, de la energía afluente actual y de la energía almacenada en el embalse equivalente en el mes actual y dos meses atrás.
Ingeniare. Revista chilena de ingeniería, 2008
El objetivo de este artículo es presentar, a aquellos investigadores neó tos, una serie de pautas... more El objetivo de este artículo es presentar, a aquellos investigadores neó tos, una serie de pautas que faciliten los procesos de escritura y publicación de manuscritos en revistas cientí cas de corriente principal. En la comunidad cientí ca, una investigación sólo se completa cuando los resultados se socializan a través de artículos en revistas arbitradas que garanticen la originalidad, importancia y validez cientí ca de los hallazgos reportados. Consecuentemente, es necesario que los investigadores desarrollen destrezas y habilidades en la escritura cientí ca. Sin embargo, el proceso de preparación de un manuscrito, desde su fase inicial hasta su publicación nal, suele ser largo y difícil, debido a algunos factores clave que tienen in uencia sobre su calidad. Este artículo suministra lineamientos generales sobre la preparación de manuscritos y entrega, también, pautas alrededor de las buenas prácticas de escritura de artículos con miras a facilitar la revisión por parte de los árbitros. En este sentido, este trabajo constituye aporte para quienes se inician en la difusión de la investigación cientí ca.
Springer Handbook of Automation, 2009
ABSTRACT This chapter contains automation information collected from many different sources aroun... more ABSTRACT This chapter contains automation information collected from many different sources around the world, among them the US Census Bureau, the International Federation of Automatic Control (IFAC), the American Automatic Control Council (AACC), the American Bankers Association, World Robotics, and many more. Section 1 introduces automation statistical data according to its application domain, such as: financial and e-commerce, industrial, healthcare, and the service industries. In Section 2 a list of worldwide automation, control, and robotics associations is provided. Section 3 provides the reader with a list of automation labs around the world and lastly, in Section 4 a list of automation related journals and publications is presented. The statistics, organizations, and journals included in this chapter open a window into the current and emerging state of automation. While attempting to be comprehensive, because of the broadness of automation, the authors recognize that there may be additional relevant items, but none were omitted intentionally. The scope of the information is meant to highlight and expose the broad span, applications, concerns and benefits of automation, and clearly, cannot completely include all areas of automation influence. Beyond this chapter, other chapters in this handbook provide additional statistical data directly related to their specific topic. For the complete review see Zbl 1218.93004.
Studies in Computational Intelligence, 2013
End users leave traces of behavior all over the Web all times. From the explicit or implicit feed... more End users leave traces of behavior all over the Web all times. From the explicit or implicit feedback of a multimedia document or a comment in an online social network, to a simple click in a relevant link in a search engine result, the information that we as users pour into the Web defines its actual representation, which is independent for each user. Our usage can be represented by different sources of data, for which different collection strategies must be considered, as well as the merging and cleaning techniques for Web usage data. Once the data is properly preprocessed, the identification of an individual user within the Web can be a complex task. Understanding the whole life of a user within a session in a Web site and the path that was pursued involves advanced data modeling and a set of assumptions which are modified every day, as new ways to interact with the online content are created. The objective is to understand the behaviour and preferences of a web user, also when several privacy issues are involved, which, as of today, are not clear how to be properly addressed. In this chapter, all previous topics regarding the processing of Web usage data are extensively discussed.
Proceedings of the first international conference on Autonomous agents - AGENTS '97, 1997
Google, Inc. (search), Subscribe (Full Service), Register (Limited Service, Free), Login. Search:... more Google, Inc. (search), Subscribe (Full Service), Register (Limited Service, Free), Login. Search: The ACM Digital Library The Guide. ...
2012 7th Colombian Computing Congress (CCC), 2012
ABSTRACT In this article, we evaluate the autoregressive neural networks (ARNN) in forecasting ti... more ABSTRACT In this article, we evaluate the autoregressive neural networks (ARNN) in forecasting time series with trend and seasonal cycle. For the time series analyzed, we found that the combined application of simple and seasonal differentiation factors does not necessarily contribute to better forecasts than if these are applied separately, and that while the preprocessing of data helps to better forecast does not mean that models are not able to capture both the trend and the seasonal cycle present in the untransformed series. In this regard, both the ARNN as the MLP (used as a benchmark) gave better results than the SARIMA process, which in the case of MLP contradicts what is stated in [15], so the assertion that MLPs do not capture these components applies only in special cases. This conclusion can be extended to ARNN model.
IEEE Latin America Transactions, 2014
2014 IEEE Congress on Evolutionary Computation (CEC), 2014
Lecture Notes in Computer Science, 2014
2010 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology, 2010
A biologically inspired cognitive model is presented for human decision making and applied to the... more A biologically inspired cognitive model is presented for human decision making and applied to the simulation of the web user. The model is based on the Neurophysiology description of multiple decision process; this is a well proven psychological theory. The model simulates the behaviour of a real user on a website and it was observed that the distribution of artificial web users in sessions successfully simulates a genuine user's web mode of behaviour. On the hypothesis that the adjusted artificial web user behaves statistically similar to the human web users, a system was created for the improvement of the structure of a web site based on stochastic simulations as a Proof of Concept. Since simulation recover observed statistical behaviour, changes on a web site are used to predict changes on navigational patterns.
2012 7th Colombian Computing Congress (CCC), 2012
ABSTRACT In the practice, to find the parameters for nonlinear Moving average(NLMA) models is dif... more ABSTRACT In the practice, to find the parameters for nonlinear Moving average(NLMA) models is difficult because there are not analytical way to solve maximum likelihood function. In this paper, we compare the parameters estimated for NLMA model with Supernova and PSO-DE algorithm. We found that Supernova estimation is better in terms of a reduction of error and nearest solutions to optimal point.
DYNA, 2014
The Committee on Publication Ethics (COPE) -conformed by the main scientific publishers-has warne... more The Committee on Publication Ethics (COPE) -conformed by the main scientific publishers-has warned about the increasing number of ethical problems in scientific publication and recent scandals seem to indicate that ethical misconduct is repetitive. Ethical problems in scientific publication arise when the person deviates from expected moral behavior. Misconduct may be explained, at least in part, because many postgraduate students and young researchers seem to understand that ethical problems are only related to plagiarism of complete works or duplication of publications, and because it seems to be a lack of knowledge of the ethical standards in scientific publication. However, there are many other aspects conducing to ethical problems. The objective of this paper is to discuss and spread the ethical position of the main scientific publishers and researchers with the aim of build a unified point of view. In this paper, seventeen tips for avoiding ethical problems in scientific publication are presented, explained and discussed. I hope that this work will be valuable for postgraduate students and young researchers and answers many common questions about ethics in scientific publication.
The Information Society, 2010
Google, Inc. (search), Subscribe (Full Service), Register (Limited Service, Free), Login. Search:... more Google, Inc. (search), Subscribe (Full Service), Register (Limited Service, Free), Login. Search: The ACM Digital Library The Guide. ...
International Journal of Production Economics, 2008
In this work, a new method to resolve conflict situations (Mcr) common in collaborative facility ... more In this work, a new method to resolve conflict situations (Mcr) common in collaborative facility design using computer support tools is developed. The method uses traditional human conflict resolution approaches that have been used successfully by others. It incorporates ...
International Journal of Computer Integrated Manufacturing, 2008
In collaborative e-work over largely distributed organisations it is often necessary to match set... more In collaborative e-work over largely distributed organisations it is often necessary to match sets of components or suppliers for better-quality assemblies, or for better-quality contracts. In such cases, the importance of efficiently and effectively sharing, matching and leveraging real-time information is demonstrated in e-work and e-manufacturing systems via a best-matching protocol (BMP). The developed BMP enables better and quicker matches for both geometrical (e.g. physical parts) and non-geometrical attributes (e.g. members in a supply network). Improvement is measured using the newly defined best-fit index. Four statistical distributions are applied to characterise part specifications of the physical components. The statistical analysis corroborates that the performance of BMP in yielding significantly better matches for geometrical specifications (within acceptable tolerance limits) is consistent across uniform and normal distributions. The economic value of a match (EVM) is also evaluated for different economic costs. The results of applying EVM indicate that organisations need to define policies for matching as it is not always economically desirable to match parts. A double-match that includes a best-matching process for non-geometrical attributes and prior matching results from the manufacturing scenario is also performed to identify suppliers that provide the best parts and costs combinations.
Ingeniare. Revista chilena de ingeniería, 2010
La predicción de series de tiempo es un importante problema de investigación debido a sus implica... more La predicción de series de tiempo es un importante problema de investigación debido a sus implicaciones en ingeniería, economía, finanzas y ciencias sociales. Un importante tópico de esta problemática es el desarrollo de nuevos modelos y su comparación con aproximaciones previas en términos de la precisión del pronóstico. Recientemente, las máquinas de vectores de soporte (SVM) han sido usadas para la predicción de series de tiempo, pero las experiencias reportadas son limitadas y hay algunos problemas relacionados con su especificación. El objetivo de este artículo es proponer una técnica novedosa para estimar algunas constantes en las SVM que usualmente son fijadas en forma empírica por el modelador. La técnica propuesta es usada para estimar varias SVM con el fin de pronosticar cinco series benchmark; los resultados obtenidos son comparados con las estadísticas reportadas en otros artículos. La metodología propuesta permite obtener SVM competitivas para las series pronosticadas en comparación con los resultados obtenidos usando otros modelos más tradicionales.
Ingeniare. Revista chilena de ingeniería, 2007
En este trabajo se propone la representación de la dinámica del Índice de Oscilación del Sur usan... more En este trabajo se propone la representación de la dinámica del Índice de Oscilación del Sur usando una nueva clase de modelo híbrido no lineal. En este nuevo modelo, la no-linealidad en la media es representada usando un sistema adaptativo neurodifuso de inferencia (ANFIS), mientras que la varianza es representada usando una componente autoregresiva heterocedástica condicional. En adición, una estrategia de especificación es desarrollada para el modelo propuesto, la cual está basada en una batería de pruebas estadísticas. Los resultados indican que el modelo propuesto captura de mejor forma las principales características de la serie de tiempo estudiada, en comparación con otros modelos competitivos.
Ingeniare. Revista chilena de ingeniería, 2008
Una red neuronal autorregresiva es estimada para el precio mensual brasileño de corto plazo de la... more Una red neuronal autorregresiva es estimada para el precio mensual brasileño de corto plazo de la electricidad, la cual describe mejor la dinámica de los precios que un modelo lineal autorregresivo y que un perceptrón multicapa clásico que usan las mismas entradas y neuronas en la capa oculta. El modelo propuesto es especificado usando un procedimiento estadístico basado en el contraste del radio de verosimilitud. El modelo pasa una batería de pruebas de diagnóstico. El procedimiento de especificación propuesto permite seleccionar el número de unidades en la capa oculta y las entradas a la red neuronal, usando pruebas estadísticas que tienen en cuenta la cantidad de los datos y el ajuste del modelo a la serie de precios. La especificación del modelo final demuestra que el precio para el próximo mes es una función no lineal del precio actual, de la energía afluente actual y de la energía almacenada en el embalse equivalente en el mes actual y dos meses atrás.
Ingeniare. Revista chilena de ingeniería, 2008
El objetivo de este artículo es presentar, a aquellos investigadores neó tos, una serie de pautas... more El objetivo de este artículo es presentar, a aquellos investigadores neó tos, una serie de pautas que faciliten los procesos de escritura y publicación de manuscritos en revistas cientí cas de corriente principal. En la comunidad cientí ca, una investigación sólo se completa cuando los resultados se socializan a través de artículos en revistas arbitradas que garanticen la originalidad, importancia y validez cientí ca de los hallazgos reportados. Consecuentemente, es necesario que los investigadores desarrollen destrezas y habilidades en la escritura cientí ca. Sin embargo, el proceso de preparación de un manuscrito, desde su fase inicial hasta su publicación nal, suele ser largo y difícil, debido a algunos factores clave que tienen in uencia sobre su calidad. Este artículo suministra lineamientos generales sobre la preparación de manuscritos y entrega, también, pautas alrededor de las buenas prácticas de escritura de artículos con miras a facilitar la revisión por parte de los árbitros. En este sentido, este trabajo constituye aporte para quienes se inician en la difusión de la investigación cientí ca.