http://sedici.unlp.edu.ar:80/handle/10915/134633 Thu, 12 Nov 2026 08:15:07 GMT 2026-11-12T08:15:07Z http://sedici.unlp.edu.ar:80/handle/10915/135316 Articulo Electronic Journal of SADIO; vol. 6 A design and development tool to achieve artificial neural networks (ANN) implemented in a Field Programmable Gate Array (FPGA), is presented in this article. Its main components and functionality are thoroughly described. This tool, called NNGen, allows constructing digital ANN, which are easily programmable selecting different parameters. The output of such programming task is directly VHDL code to be ported to, in principle, any chip of the mentioned kind. A case study of a multilayer perceptron applied to weather forecast that was fully designed with NNGen, is also analyzed to obtain some conclusions. Sat, 26 Jun 2004 00:00:00 GMT http://sedici.unlp.edu.ar:80/handle/10915/135316 2004-06-26T00:00:00Z A design and development tool to achieve artificial neural networks (ANN) implemented in a Field Programmable Gate Array (FPGA), is presented in this article. Its main components and functionality are thoroughly described. This tool, called NNGen, allows constructing digital ANN, which are easily programmable selecting different parameters. The output of such programming task is directly VHDL code to be ported to, in principle, any chip of the mentioned kind. A case study of a multilayer perceptron applied to weather forecast that was fully designed with NNGen, is also analyzed to obtain some conclusions. http://sedici.unlp.edu.ar:80/handle/10915/135315 Articulo Electronic Journal of SADIO; vol. 6 A Multi-Objective Genetic Algorithm (MOGA) application, which is based on the aggregating approach, is proposed in this article. Its aim is to find a consistent instrument configuration for industrial process plants that will constitute a convenient initial set of input data for structural Observability Analysis Algorithms (OAs). The better this configuration is, the faster the OAs will converge to a satisfactory solution. Algorithmic effectiveness was evaluated through the analysis of small academic case studies. The results obtained through our algorithm show excellent performance. Therefore, it can be stated that the prototype presented in this work is good enough to serve as a sound basis for the development of the definitive MOGA module, whose implementation will support large-size industrial plant models. Sat, 26 Jun 2004 00:00:00 GMT http://sedici.unlp.edu.ar:80/handle/10915/135315 2004-06-26T00:00:00Z A Multi-Objective Genetic Algorithm (MOGA) application, which is based on the aggregating approach, is proposed in this article. Its aim is to find a consistent instrument configuration for industrial process plants that will constitute a convenient initial set of input data for structural Observability Analysis Algorithms (OAs). The better this configuration is, the faster the OAs will converge to a satisfactory solution. Algorithmic effectiveness was evaluated through the analysis of small academic case studies. The results obtained through our algorithm show excellent performance. Therefore, it can be stated that the prototype presented in this work is good enough to serve as a sound basis for the development of the definitive MOGA module, whose implementation will support large-size industrial plant models. http://sedici.unlp.edu.ar:80/handle/10915/135314 Articulo Electronic Journal of SADIO; vol. 6 The search for patterns in data sets is a fundamental task in Data Mining, where Machine Learning algorithms are generally used. However, Machine Learning algorithms have biases that strengthen the classifica-tion task, not taking into consideration exceptions. Exceptions contra-dict common sense rules. They are generally unknown, unexpected and contradictory to the user believes. For this reason, exceptions may be interesting. In this work we propose a method to find exceptions out from common sense rules. Besides, we apply the proposed method in a real world data set, to discover rules and exceptions in the HIV virus protein cleavage process. Sat, 26 Jun 2004 00:00:00 GMT http://sedici.unlp.edu.ar:80/handle/10915/135314 2004-06-26T00:00:00Z The search for patterns in data sets is a fundamental task in Data Mining, where Machine Learning algorithms are generally used. However, Machine Learning algorithms have biases that strengthen the classifica-tion task, not taking into consideration exceptions. Exceptions contra-dict common sense rules. They are generally unknown, unexpected and contradictory to the user believes. For this reason, exceptions may be interesting. In this work we propose a method to find exceptions out from common sense rules. Besides, we apply the proposed method in a real world data set, to discover rules and exceptions in the HIV virus protein cleavage process. http://sedici.unlp.edu.ar:80/handle/10915/135313 Articulo Electronic Journal of SADIO; vol. 6 In this paper we define a non-continuous discrete dynamical system re-lated to a transfer scheme designed originally to approximate imputations in the core of balanced games. We show that the dynamical system may have either periodic point of period 1 (fixed points) or periodic points with period greater than one, but not both. Moreover, the fixed points of the dynamical system characterize the core of a balanced game. On the other side, periodic points of period greater than one are associated with certain class of cycles of pre-imputations that can appear in non-balanced games (maximal U-cycles). For monotonic non-balanced 3-person games we de-scribe completely the set of periodic points and their associated (forward) stable sets. Sat, 26 Jun 2004 00:00:00 GMT http://sedici.unlp.edu.ar:80/handle/10915/135313 2004-06-26T00:00:00Z In this paper we define a non-continuous discrete dynamical system re-lated to a transfer scheme designed originally to approximate imputations in the core of balanced games. We show that the dynamical system may have either periodic point of period 1 (fixed points) or periodic points with period greater than one, but not both. Moreover, the fixed points of the dynamical system characterize the core of a balanced game. On the other side, periodic points of period greater than one are associated with certain class of cycles of pre-imputations that can appear in non-balanced games (maximal U-cycles). For monotonic non-balanced 3-person games we de-scribe completely the set of periodic points and their associated (forward) stable sets. http://sedici.unlp.edu.ar:80/handle/10915/135312 Contribucion a revista Editorial note by Gerardo Acosta Electronic Journal of SADIO; vol. 6 Last September, in occasion of the fifth Argentinean Symposium on Artificial Intelligence (V ASAI), all of the attendants had the satisfaction to see the great evolution of the discipline in this region. This growth, since these techniques arose as effective problem solving methods for every day life in the first half of the past century, had not a direct correlation in our country since the creation of this event within the SADIO’s Congress. From the first ASAI in 1999, up to nowadays, a good quantity of valuable activity in AI can be shared among researchers and entrepreneurs, in a high level forum, with a pleasant climate. In this fifth ASAI, 26 papers were accepted for presentation, after a thorough refereeing process. Three reviewers from a high-qualified international Scientific Committee carefully evaluated each work. These works were punctuated taking into account its originality, presentation, relevance of the results, and general quality. The five ones with the highest scores were selected to be published in this special issue of the Electric Journal of SADIO. Then, in this volume, we will be able to enjoy the paper from the team of Carolina Monard at the University of Sao Paulo@Sao Carlos, Brazil, A Method for Refining Knowledge Rules Using Exceptions, in which they present a contribution to the data mining problem with the knowledge-based approach. In this particular task, to manage exceptions is not at all a trivial difficulty. The solution proposed was tested upon real data in the HIV virus protein cleavage process. Sat, 26 Jun 2004 00:00:00 GMT http://sedici.unlp.edu.ar:80/handle/10915/135312 2004-06-26T00:00:00Z Last September, in occasion of the fifth Argentinean Symposium on Artificial Intelligence (V ASAI), all of the attendants had the satisfaction to see the great evolution of the discipline in this region. This growth, since these techniques arose as effective problem solving methods for every day life in the first half of the past century, had not a direct correlation in our country since the creation of this event within the SADIO’s Congress. From the first ASAI in 1999, up to nowadays, a good quantity of valuable activity in AI can be shared among researchers and entrepreneurs, in a high level forum, with a pleasant climate. In this fifth ASAI, 26 papers were accepted for presentation, after a thorough refereeing process. Three reviewers from a high-qualified international Scientific Committee carefully evaluated each work. These works were punctuated taking into account its originality, presentation, relevance of the results, and general quality. The five ones with the highest scores were selected to be published in this special issue of the Electric Journal of SADIO. Then, in this volume, we will be able to enjoy the paper from the team of Carolina Monard at the University of Sao Paulo@Sao Carlos, Brazil, A Method for Refining Knowledge Rules Using Exceptions, in which they present a contribution to the data mining problem with the knowledge-based approach. In this particular task, to manage exceptions is not at all a trivial difficulty. The solution proposed was tested upon real data in the HIV virus protein cleavage process. http://sedici.unlp.edu.ar:80/handle/10915/135310 Articulo Electronic Journal of SADIO; vol. 6 In this paper, a parameter estimation problem for a Takagi-Sugeno fuzzy dynamic system is formulated under the assumption that the premises in the membership functions are known. A linear expression in consequent parameters is obtained under this assumption. If the system is time-varying, the parameters can be determined by recursive estimation techniques. As an alternate approach, the use of multi-resolution wavelets is proposed. Furthermore, a parameter estimation toolbox for fuzzy dynamic models is developed which is then applied to a simple example and to the Mackey-Glass chaotic time series. Sat, 26 Jun 2004 00:00:00 GMT http://sedici.unlp.edu.ar:80/handle/10915/135310 2004-06-26T00:00:00Z In this paper, a parameter estimation problem for a Takagi-Sugeno fuzzy dynamic system is formulated under the assumption that the premises in the membership functions are known. A linear expression in consequent parameters is obtained under this assumption. If the system is time-varying, the parameters can be determined by recursive estimation techniques. As an alternate approach, the use of multi-resolution wavelets is proposed. Furthermore, a parameter estimation toolbox for fuzzy dynamic models is developed which is then applied to a simple example and to the Mackey-Glass chaotic time series. http://sedici.unlp.edu.ar:80/handle/10915/135309 Articulo Electronic Journal of SADIO; vol. 6 Interface agents are computer programs that learn users' preferences to provide them personalized assistance with their computer-based tasks. In order to personalize the interaction with users, interface agents must learn how to best interact with each user and how to provide them assistance of the right sort at the right time. Particularly, an interface agent has to discover when the user needs a suggestion to solve a problem, when he requires only a warning about it, when he wants the agent to execute an action and when he wants the agent to do just nothing. In this work we propose a learning algorithm, named WATSON, to tackle this problem. The WATSON algorithm enables an interface agent to adapt its behavior and its interaction with a user to the user's assistance requirements. Our algorithm uses association rules (AR) to discover associations among problem situations and a user's assistance requirements in a given application domain. Sat, 26 Jun 2004 00:00:00 GMT http://sedici.unlp.edu.ar:80/handle/10915/135309 2004-06-26T00:00:00Z Interface agents are computer programs that learn users' preferences to provide them personalized assistance with their computer-based tasks. In order to personalize the interaction with users, interface agents must learn how to best interact with each user and how to provide them assistance of the right sort at the right time. Particularly, an interface agent has to discover when the user needs a suggestion to solve a problem, when he requires only a warning about it, when he wants the agent to execute an action and when he wants the agent to do just nothing. In this work we propose a learning algorithm, named WATSON, to tackle this problem. The WATSON algorithm enables an interface agent to adapt its behavior and its interaction with a user to the user's assistance requirements. Our algorithm uses association rules (AR) to discover associations among problem situations and a user's assistance requirements in a given application domain.