Autonomous mobile robots require efficient control of their movement. This paper
focuses on a pro... more Autonomous mobile robots require efficient control of their movement. This paper focuses on a prototype implementing a consciousness model for autonomous mobile robots. The prototype is named MARCOS (Movement of Autonomous Robotics Codelet System). Here is an in-depth analysis of this secondary controller that complements the main close-loop controller providing complex information about the current robot status, and suggesting alternative actions. This paper describes the MARCOS prototype and its conscious subsystem, as well as statistics on how Concepts can model real-world indoor environments.
Las piezas audiovisuales ayudan a comprender los contenidos complejos dando respuesta a este tipo... more Las piezas audiovisuales ayudan a comprender los contenidos complejos dando respuesta a este tipo de interrogantes que explican de forma corta, sencilla y efectiva, la historia, aplicaciones y funcionamiento de dichas carreras profesionales. A partir de esta necesidad de crear determinados lineamientos del lenguaje audiovisual para la enseñanza de las diferentes ingenierías que actualmente existen a nivel global, surge el proyecto MIDA para entrenar a estudiantes de Ingeniería en tópicos de alta exigencia de abstracción, y LEARNITRON es un subproyecto enfocado al estudio y aplicación de las técnicas audiovisuales en este contexto. En el presente trabajo se instaura una breve descripción acerca de las técnicas apropiadas a la hora de realizar vídeos de contenido educativo, alcances de éste tipo de herramientas audiovisuales y finalizando con la presentación de pruebas estadísticas que sustentan el incentivo de este tipo de técnicas en la enseñanzas. Dicho proyecto propone una innovadora metodología de refinado y mejora del proceso de aprendizaje ingenieriles.
The goal of this paper is to define a proper audio protocol for ASD patients. As part of the scop... more The goal of this paper is to define a proper audio protocol for ASD patients. As part of the scope sheets collection protocol, a global and a case-based evaluation methodology are presented. There is also a kinetic pattern modeling in relation to sound stimuli precisely defined. Findings here indicate that the proposed protocol is good for proper data collection required to biomechanical and behavioral processing. It also enable to automate certain subtle information usually collected just by expert evaluation. Thereby it is possible to evaluate relevant performance in patients with an automatic system.
Certain subjects in Engineering cover complex concepts.
Multimedia may provide a simple, short an... more Certain subjects in Engineering cover complex concepts. Multimedia may provide a simple, short and effective way to explain them. This paper aims to describe a virtual learning environment named MIDA. It has been designed to train and evaluate a set of tips to improve teaching in that area. LEARNITRON is an offspring of the project. It is a sub project for automatic modelling learning profile of students according to the combination of students profile, topics to be taught and current set of multimedia tools. The goal of this paper is to sketch LEARNITRON and how to improve multimedia tools for teaching, based on basic machine learning techniques. The resulting model may be useful to define specific features for better educational videos, and to acknowledge the right scope of multimedia as a way to teach or train students in Engineering subjects based not on personal opinion but on statistical findings
Vehicle and pedestrian risks can be modeled in order to advice drivers
and persons. A good model... more Vehicle and pedestrian risks can be modeled in order to advice drivers and persons. A good model requires the ability to adapt itself to several environ mental variations and to preserve essential information about the area under scope. This paper aims to present a proposal based on a Machine Learning extension for timing named Harmonic Systems. A global description of the problem, its rele vance and status of the field are also included
There are several approaches to interact with ASD patients.
Many of them don't relay on vocal abi... more There are several approaches to interact with ASD patients. Many of them don't relay on vocal abilities because of the wide spectrum of symptoms. But taking data from ML perspective may introduce new sources of information for a better interpretation of patient's behaviour and helps build new parameters to model individual performance. This paper presents some initial findings in that direction based on the automatic evaluation of real cases as a way to build a parameterised description of patients. Preliminary analysis in patients suggest that the individual performance may be compared with the universe of patients, it is possible to elaborate a performance profiling and to model evolution profiles that are present in a predefined universe of patients. The team is currently defining and testing new audio and video protocols to collect additional parameters. From these findings, it is possible to build a parametric reasoning model using MLW.
Information management is currently one of the topics with most impact. Processes are studied in ... more Information management is currently one of the topics with most impact. Processes are studied in this paper in their temporality, mainly to be able to make predictions. Interesting information arises when what is being examined is not the complex information produced along the development of a process, but the temporality itself for the process. Such an alternative point of view enables the relation of this analysis to a real process, and its application in the very moment the events are taking place, knowing that there is a starting point of certain a priori knowledge. This work is shows this alternative stance, from the temporal perspective. There are applications of concepts and analogies brought from a different environment where time is handled in a more natural way. We shall call this approach "Harmonic Systems". We shall present the basis for harmonic systems, the focused approach we propose as a type of time mining, and its application
This work introduces a software model for signature validation. The proposed technique is
based o... more This work introduces a software model for signature validation. The proposed technique is based on wavelets for image processing and an artificial neural network for data classification. Handwritten Signature Validation (HSV) is a proven and accepted procedure for identity validation by means of non-invasive and non-intrusive procedures (features which do biometrics controversial as an authentication method). The experimental prototype is built on software developed by AI Group Research Laboratories at Universidad de Palermo, project named PI-1 for handwritten signature recognition. The aim of the model is to improve automatic HSV reliability through digital imaging algorithms and computational intelligence techniques. In the next sections, a brief presentation of the work.
Autonomous mobile robots require efficient control of their movement. This paper focuses on a pro... more Autonomous mobile robots require efficient control of their movement. This paper focuses on a prototype implementing a consciousness model for autonomous mobile robots. The prototype is named FIC (Fluent Interactive Codelets framework). Here is an in-depth analysis of this secondary controller that complements the main close-loop controller providing complex information about the current robot status, and suggesting alternative actions. This paper describes the FIC prototype and its conscious subsystem, as well as statistics on how Concepts can model real-world indoor environments.Autonomous mobile robots require efficient control of their movement. This paper focuses on a prototype implementing a consciousness model for autonomous mobile robots. The prototype is named FIC (Fluent Interactive Codelets framework). Here is an in-depth analysis of this secondary controller that complements the main close-loop controller providing complex information about the current robot status, and suggesting alternative actions. This paper describes the FIC prototype and its conscious subsystem, as well as statistics on how Concepts can model real-world indoor environments.
The aim of this paper is to implement metrics and to define indicators to provide a unified crite... more The aim of this paper is to implement metrics and to define indicators to provide a unified criteria for evaluation method of the performance of autonomous mobile robots using different control algorithms. There is a first description of the mobile robot problem and the importance of a standardized process for evaluation of robot performances. There is a comparison between a simple navigation controller and an intelligent prototype based on consciousness. The architecture main features are also outlined. Test cas
Different articles and surveys on the matter identify a remarkable success of robotics on static ... more Different articles and surveys on the matter identify a remarkable success of robotics on static environments that has not replicated on dynamic areas of application. This paper presents the proposal of a research project that identifies the main hurdles of this issue and present feasible solutions for the matter.
The aim of this paper is to introduce a new model of Decision Support System (DDS) for transporta... more The aim of this paper is to introduce a new model of Decision Support System (DDS) for transportation and infrastructure enterprise logistics. The model evaluates different parameters relevant in the field. They relate to micro and macro economic factors that somehow could affect the activity. It also evaluates other variables related to political, socio-cultural, technological and local regulatory topics. The model is being under implementation as HERCULES prototype. The architecture is centered on an Expert System (ES) that evaluates the best Data Mining approaches for the current problem and context, allowing a proper forecasting of potential scenarios. It customizes and feeds information to a second ES, a Meta-ES. The first ES provides the second the context and expertise of a Data Mining specialist, whilst the second ES has specific expertise in the field. It advises the key decision makers with the appropriate information based on the risks faced as part of a specific decision. The contribution of this prototype are the original variables definition and treatment and the auto-configuration of the Meta-Expert System.
This paper presents what the authors consider to be a new distributed memory architecture for sel... more This paper presents what the authors consider to be a new distributed memory architecture for self-organizing distributed systems. The work in progress presented here focuses on the representation of a distributed memory for multi-robot systems. It states the main characteristics of a shared environment and provides suitable interfaces to allow autonomous mobile robots to query, publish and exchange relevant information without requiring a central data storage. The memory is based on a combination of fuzzy systems, distributed Self- Organizing Maps (SOM), a new specific design information handshake between robots and a new model for approach for world-map global administration.
This paper describes briefly a model implemented in the conversational robot named WIH (Word Inte... more This paper describes briefly a model implemented in the conversational robot named WIH (Word Intelligent Handler) and several statistics showing how it represents information extracted from Spanish dialogs. In order to get knowledge quickly, dialogs are processed with a batch interface. The prototype has complex architecture with many components. It is organized in several memories; each one models Spanish sentences that implement Morphosintactic Linguistic Wavelets (MLW). Using a knowledge-representation mechanism, WIH can model answers automatically taking previous dialogs with or between humans. Along with test cases description are also statistical analyses of the results obtained with real dialogs. It is shown here that WIH prototype handles structures called E ce that are the kernel of its learning in a human-like fashion, and that it is able to use it as a model for answers in future dialogs.
A partir de la explosión de Internet nos encontramos en un entorno que abunda en información. En... more A partir de la explosión de Internet nos encontramos en un entorno que abunda en información. En este contexto han tomado relevancia las herramientas que permiten obtener los sitios web de mayor importancia en base a palabras clave, siendo hoy un buscador la página más importante de Internet. Los buscadores son herramientas que se encargan de agrupar y clasificar páginas para que a la hora de buscar información sobre un tema específico se obtenga un listado priorizado de aquellas con la información buscada. Una importante tarea la llevan adelante los Crawlers, que se encargan de explorar la Web de forma automatizada para ir indexando nuevos sitios en base a diversas métricas, alimentando así la base de conocimiento del buscador. En este trabajo se exploran los componentes y la viabilidad de una forma de exploración basada en Clustering que toma en cuenta la distancia semántica transponiendo el “conocimiento” en un espacio multidimensional. Esta es la motivación principal del proyecto GDARIM. A su vez, GDARIM es la fuente de recursos de un sistema de Opinion Mining, también desarrollado por los autore
The aim of this paper is to introduce a Decision
Support System (DDS) to be applied within transp... more The aim of this paper is to introduce a Decision Support System (DDS) to be applied within transportation and infrastructure enterprise level. The prototype retrieves from different sources a set of parameters relative to the sector, associated with micro and macro economic factors which directly affect the activity, complemented with other studies related to political, socio-cultural, technological and local regulatory parameters. The architecture is centered on an Expert System which provides the appropriate data mining tools set to be applied for forecasting potential scenarios, while allowing the configuration of the Meta-Expert Inferential System. The main findings of the first system is intended to provide the second the context and expertise of a data mining Specialist, while the second system has the specific expertise allowing to advise the key decision makers with the appropriate information based on the risks to be faced in a particular business. The originality of this prototype lies in the variables definition ever used in the sector and the configuration of the Meta-Expert System.
El presente trabajo presenta un sistema de ayuda a la toma de decisiones, conocido como DSS (Deci... more El presente trabajo presenta un sistema de ayuda a la toma de decisiones, conocido como DSS (Decision Support System), aplicado al ámbito de las empresas de infraestructura vial y tránsito. El prototipo posee desarrollado internamente una base de parámetros propios del sector, asociados a los factores macro económicos que afectan en forma directa a la actividad, complementados con el estudio del contexto político, socio-cultural, tecnológico y regulatorio local.
La arquitectura se centra en un Sistema Experto que establece el conjunto apropiado de herramientas de Minería de Datos a aplicar para una correcta predicción de escenarios posibles, y configura el sistema inferencial de un Meta-Sistema Experto, al cual provee también de un conjunto de datos específicos para aplicar su conocimiento experto. Los eventuales hallazgos del primer sistema tienen como objetivo ofrecer al segundo sistema el contexto y expertise de un especialista en minería de datos, mientras que el segundo posee el conocimiento experto específico que le permite asesorar a los responsables de la toma de decisión la información apropiada para la toma de decisiones en base a los riesgos que deben afronta en un determinado negocio.
En los años 50 los sistemas inteligentes eran
poco más que una tecnología de guerra que
buscaba r... more En los años 50 los sistemas inteligentes eran poco más que una tecnología de guerra que buscaba reubicarse en la plaza científica y tecnológica. El furioso éxito de los primeros Sistemas Expertos (sistemas informáticos que imitan a un experto humano en un área específica) y algunas variantes de Redes Neuronales artificiales (especialmente aquellas que permitían reconocimientos de patrones) durante los años 60 culminó en un colapso casi total luego de que la comunidad verificara en los 80 que aquellas tecnologías distaban mucho de satisfacer las exigentes expectativas de ese entonces.
This study evaluates a model for signature validation. The proposed technique is based on wavelet... more This study evaluates a model for signature validation. The proposed technique is based on wavelets for image processing and an Artificial Neural Network for data classification. Handwritten Signature Validation (HSV) is a proven and accepted procedure for identity validation by means of non-invasive and non- intrusive procedures. The experimental prototype is built on software developed by AI Group Research Laboratories at Universidad de Palermo, project named PI-1 for handwritten signature validation. The aim of the research is to improve automatic HSV reliability through digital imaging algorithms and computational intelligence techniques. In the next sections, a brief presentation of the study, essays characteristics, results and a solution design.
Since its launch, Internet has become a vast source of information in Natural Language. Accessing... more Since its launch, Internet has become a vast source of information in Natural Language. Accessing such information corpus efficiently is a matter of discussion and research in various fields. Search engines are in charge of grouping and classifying web pages in terms of keywords. Web Crawler processes perform such tasks. During exploration the websites are indexed according to different metrics, feeding the knowledge base of the search engine. Exploring hypertext and its links presents an exponential order of growth in terms of computational time. This paper presents the GDARIM project a Web Crawler that presents techniques that improve the accuracy of the information retrieved on a crawling process for a given set of keywords. GDARIM implements an exploration method based on Fuzzy Clustering, which takes into consideration the semantic distance of terms transposing knowledge into a multidimensional space. GDARIM is also the data source input for an Opinion Mining system that needs preprocessed information from preselected sources.
Autonomous mobile robots require efficient control of their movement. This paper
focuses on a pro... more Autonomous mobile robots require efficient control of their movement. This paper focuses on a prototype implementing a consciousness model for autonomous mobile robots. The prototype is named MARCOS (Movement of Autonomous Robotics Codelet System). Here is an in-depth analysis of this secondary controller that complements the main close-loop controller providing complex information about the current robot status, and suggesting alternative actions. This paper describes the MARCOS prototype and its conscious subsystem, as well as statistics on how Concepts can model real-world indoor environments.
Las piezas audiovisuales ayudan a comprender los contenidos complejos dando respuesta a este tipo... more Las piezas audiovisuales ayudan a comprender los contenidos complejos dando respuesta a este tipo de interrogantes que explican de forma corta, sencilla y efectiva, la historia, aplicaciones y funcionamiento de dichas carreras profesionales. A partir de esta necesidad de crear determinados lineamientos del lenguaje audiovisual para la enseñanza de las diferentes ingenierías que actualmente existen a nivel global, surge el proyecto MIDA para entrenar a estudiantes de Ingeniería en tópicos de alta exigencia de abstracción, y LEARNITRON es un subproyecto enfocado al estudio y aplicación de las técnicas audiovisuales en este contexto. En el presente trabajo se instaura una breve descripción acerca de las técnicas apropiadas a la hora de realizar vídeos de contenido educativo, alcances de éste tipo de herramientas audiovisuales y finalizando con la presentación de pruebas estadísticas que sustentan el incentivo de este tipo de técnicas en la enseñanzas. Dicho proyecto propone una innovadora metodología de refinado y mejora del proceso de aprendizaje ingenieriles.
The goal of this paper is to define a proper audio protocol for ASD patients. As part of the scop... more The goal of this paper is to define a proper audio protocol for ASD patients. As part of the scope sheets collection protocol, a global and a case-based evaluation methodology are presented. There is also a kinetic pattern modeling in relation to sound stimuli precisely defined. Findings here indicate that the proposed protocol is good for proper data collection required to biomechanical and behavioral processing. It also enable to automate certain subtle information usually collected just by expert evaluation. Thereby it is possible to evaluate relevant performance in patients with an automatic system.
Certain subjects in Engineering cover complex concepts.
Multimedia may provide a simple, short an... more Certain subjects in Engineering cover complex concepts. Multimedia may provide a simple, short and effective way to explain them. This paper aims to describe a virtual learning environment named MIDA. It has been designed to train and evaluate a set of tips to improve teaching in that area. LEARNITRON is an offspring of the project. It is a sub project for automatic modelling learning profile of students according to the combination of students profile, topics to be taught and current set of multimedia tools. The goal of this paper is to sketch LEARNITRON and how to improve multimedia tools for teaching, based on basic machine learning techniques. The resulting model may be useful to define specific features for better educational videos, and to acknowledge the right scope of multimedia as a way to teach or train students in Engineering subjects based not on personal opinion but on statistical findings
Vehicle and pedestrian risks can be modeled in order to advice drivers
and persons. A good model... more Vehicle and pedestrian risks can be modeled in order to advice drivers and persons. A good model requires the ability to adapt itself to several environ mental variations and to preserve essential information about the area under scope. This paper aims to present a proposal based on a Machine Learning extension for timing named Harmonic Systems. A global description of the problem, its rele vance and status of the field are also included
There are several approaches to interact with ASD patients.
Many of them don't relay on vocal abi... more There are several approaches to interact with ASD patients. Many of them don't relay on vocal abilities because of the wide spectrum of symptoms. But taking data from ML perspective may introduce new sources of information for a better interpretation of patient's behaviour and helps build new parameters to model individual performance. This paper presents some initial findings in that direction based on the automatic evaluation of real cases as a way to build a parameterised description of patients. Preliminary analysis in patients suggest that the individual performance may be compared with the universe of patients, it is possible to elaborate a performance profiling and to model evolution profiles that are present in a predefined universe of patients. The team is currently defining and testing new audio and video protocols to collect additional parameters. From these findings, it is possible to build a parametric reasoning model using MLW.
Information management is currently one of the topics with most impact. Processes are studied in ... more Information management is currently one of the topics with most impact. Processes are studied in this paper in their temporality, mainly to be able to make predictions. Interesting information arises when what is being examined is not the complex information produced along the development of a process, but the temporality itself for the process. Such an alternative point of view enables the relation of this analysis to a real process, and its application in the very moment the events are taking place, knowing that there is a starting point of certain a priori knowledge. This work is shows this alternative stance, from the temporal perspective. There are applications of concepts and analogies brought from a different environment where time is handled in a more natural way. We shall call this approach "Harmonic Systems". We shall present the basis for harmonic systems, the focused approach we propose as a type of time mining, and its application
This work introduces a software model for signature validation. The proposed technique is
based o... more This work introduces a software model for signature validation. The proposed technique is based on wavelets for image processing and an artificial neural network for data classification. Handwritten Signature Validation (HSV) is a proven and accepted procedure for identity validation by means of non-invasive and non-intrusive procedures (features which do biometrics controversial as an authentication method). The experimental prototype is built on software developed by AI Group Research Laboratories at Universidad de Palermo, project named PI-1 for handwritten signature recognition. The aim of the model is to improve automatic HSV reliability through digital imaging algorithms and computational intelligence techniques. In the next sections, a brief presentation of the work.
Autonomous mobile robots require efficient control of their movement. This paper focuses on a pro... more Autonomous mobile robots require efficient control of their movement. This paper focuses on a prototype implementing a consciousness model for autonomous mobile robots. The prototype is named FIC (Fluent Interactive Codelets framework). Here is an in-depth analysis of this secondary controller that complements the main close-loop controller providing complex information about the current robot status, and suggesting alternative actions. This paper describes the FIC prototype and its conscious subsystem, as well as statistics on how Concepts can model real-world indoor environments.Autonomous mobile robots require efficient control of their movement. This paper focuses on a prototype implementing a consciousness model for autonomous mobile robots. The prototype is named FIC (Fluent Interactive Codelets framework). Here is an in-depth analysis of this secondary controller that complements the main close-loop controller providing complex information about the current robot status, and suggesting alternative actions. This paper describes the FIC prototype and its conscious subsystem, as well as statistics on how Concepts can model real-world indoor environments.
The aim of this paper is to implement metrics and to define indicators to provide a unified crite... more The aim of this paper is to implement metrics and to define indicators to provide a unified criteria for evaluation method of the performance of autonomous mobile robots using different control algorithms. There is a first description of the mobile robot problem and the importance of a standardized process for evaluation of robot performances. There is a comparison between a simple navigation controller and an intelligent prototype based on consciousness. The architecture main features are also outlined. Test cas
Different articles and surveys on the matter identify a remarkable success of robotics on static ... more Different articles and surveys on the matter identify a remarkable success of robotics on static environments that has not replicated on dynamic areas of application. This paper presents the proposal of a research project that identifies the main hurdles of this issue and present feasible solutions for the matter.
The aim of this paper is to introduce a new model of Decision Support System (DDS) for transporta... more The aim of this paper is to introduce a new model of Decision Support System (DDS) for transportation and infrastructure enterprise logistics. The model evaluates different parameters relevant in the field. They relate to micro and macro economic factors that somehow could affect the activity. It also evaluates other variables related to political, socio-cultural, technological and local regulatory topics. The model is being under implementation as HERCULES prototype. The architecture is centered on an Expert System (ES) that evaluates the best Data Mining approaches for the current problem and context, allowing a proper forecasting of potential scenarios. It customizes and feeds information to a second ES, a Meta-ES. The first ES provides the second the context and expertise of a Data Mining specialist, whilst the second ES has specific expertise in the field. It advises the key decision makers with the appropriate information based on the risks faced as part of a specific decision. The contribution of this prototype are the original variables definition and treatment and the auto-configuration of the Meta-Expert System.
This paper presents what the authors consider to be a new distributed memory architecture for sel... more This paper presents what the authors consider to be a new distributed memory architecture for self-organizing distributed systems. The work in progress presented here focuses on the representation of a distributed memory for multi-robot systems. It states the main characteristics of a shared environment and provides suitable interfaces to allow autonomous mobile robots to query, publish and exchange relevant information without requiring a central data storage. The memory is based on a combination of fuzzy systems, distributed Self- Organizing Maps (SOM), a new specific design information handshake between robots and a new model for approach for world-map global administration.
This paper describes briefly a model implemented in the conversational robot named WIH (Word Inte... more This paper describes briefly a model implemented in the conversational robot named WIH (Word Intelligent Handler) and several statistics showing how it represents information extracted from Spanish dialogs. In order to get knowledge quickly, dialogs are processed with a batch interface. The prototype has complex architecture with many components. It is organized in several memories; each one models Spanish sentences that implement Morphosintactic Linguistic Wavelets (MLW). Using a knowledge-representation mechanism, WIH can model answers automatically taking previous dialogs with or between humans. Along with test cases description are also statistical analyses of the results obtained with real dialogs. It is shown here that WIH prototype handles structures called E ce that are the kernel of its learning in a human-like fashion, and that it is able to use it as a model for answers in future dialogs.
A partir de la explosión de Internet nos encontramos en un entorno que abunda en información. En... more A partir de la explosión de Internet nos encontramos en un entorno que abunda en información. En este contexto han tomado relevancia las herramientas que permiten obtener los sitios web de mayor importancia en base a palabras clave, siendo hoy un buscador la página más importante de Internet. Los buscadores son herramientas que se encargan de agrupar y clasificar páginas para que a la hora de buscar información sobre un tema específico se obtenga un listado priorizado de aquellas con la información buscada. Una importante tarea la llevan adelante los Crawlers, que se encargan de explorar la Web de forma automatizada para ir indexando nuevos sitios en base a diversas métricas, alimentando así la base de conocimiento del buscador. En este trabajo se exploran los componentes y la viabilidad de una forma de exploración basada en Clustering que toma en cuenta la distancia semántica transponiendo el “conocimiento” en un espacio multidimensional. Esta es la motivación principal del proyecto GDARIM. A su vez, GDARIM es la fuente de recursos de un sistema de Opinion Mining, también desarrollado por los autore
The aim of this paper is to introduce a Decision
Support System (DDS) to be applied within transp... more The aim of this paper is to introduce a Decision Support System (DDS) to be applied within transportation and infrastructure enterprise level. The prototype retrieves from different sources a set of parameters relative to the sector, associated with micro and macro economic factors which directly affect the activity, complemented with other studies related to political, socio-cultural, technological and local regulatory parameters. The architecture is centered on an Expert System which provides the appropriate data mining tools set to be applied for forecasting potential scenarios, while allowing the configuration of the Meta-Expert Inferential System. The main findings of the first system is intended to provide the second the context and expertise of a data mining Specialist, while the second system has the specific expertise allowing to advise the key decision makers with the appropriate information based on the risks to be faced in a particular business. The originality of this prototype lies in the variables definition ever used in the sector and the configuration of the Meta-Expert System.
El presente trabajo presenta un sistema de ayuda a la toma de decisiones, conocido como DSS (Deci... more El presente trabajo presenta un sistema de ayuda a la toma de decisiones, conocido como DSS (Decision Support System), aplicado al ámbito de las empresas de infraestructura vial y tránsito. El prototipo posee desarrollado internamente una base de parámetros propios del sector, asociados a los factores macro económicos que afectan en forma directa a la actividad, complementados con el estudio del contexto político, socio-cultural, tecnológico y regulatorio local.
La arquitectura se centra en un Sistema Experto que establece el conjunto apropiado de herramientas de Minería de Datos a aplicar para una correcta predicción de escenarios posibles, y configura el sistema inferencial de un Meta-Sistema Experto, al cual provee también de un conjunto de datos específicos para aplicar su conocimiento experto. Los eventuales hallazgos del primer sistema tienen como objetivo ofrecer al segundo sistema el contexto y expertise de un especialista en minería de datos, mientras que el segundo posee el conocimiento experto específico que le permite asesorar a los responsables de la toma de decisión la información apropiada para la toma de decisiones en base a los riesgos que deben afronta en un determinado negocio.
En los años 50 los sistemas inteligentes eran
poco más que una tecnología de guerra que
buscaba r... more En los años 50 los sistemas inteligentes eran poco más que una tecnología de guerra que buscaba reubicarse en la plaza científica y tecnológica. El furioso éxito de los primeros Sistemas Expertos (sistemas informáticos que imitan a un experto humano en un área específica) y algunas variantes de Redes Neuronales artificiales (especialmente aquellas que permitían reconocimientos de patrones) durante los años 60 culminó en un colapso casi total luego de que la comunidad verificara en los 80 que aquellas tecnologías distaban mucho de satisfacer las exigentes expectativas de ese entonces.
This study evaluates a model for signature validation. The proposed technique is based on wavelet... more This study evaluates a model for signature validation. The proposed technique is based on wavelets for image processing and an Artificial Neural Network for data classification. Handwritten Signature Validation (HSV) is a proven and accepted procedure for identity validation by means of non-invasive and non- intrusive procedures. The experimental prototype is built on software developed by AI Group Research Laboratories at Universidad de Palermo, project named PI-1 for handwritten signature validation. The aim of the research is to improve automatic HSV reliability through digital imaging algorithms and computational intelligence techniques. In the next sections, a brief presentation of the study, essays characteristics, results and a solution design.
Since its launch, Internet has become a vast source of information in Natural Language. Accessing... more Since its launch, Internet has become a vast source of information in Natural Language. Accessing such information corpus efficiently is a matter of discussion and research in various fields. Search engines are in charge of grouping and classifying web pages in terms of keywords. Web Crawler processes perform such tasks. During exploration the websites are indexed according to different metrics, feeding the knowledge base of the search engine. Exploring hypertext and its links presents an exponential order of growth in terms of computational time. This paper presents the GDARIM project a Web Crawler that presents techniques that improve the accuracy of the information retrieved on a crawling process for a given set of keywords. GDARIM implements an exploration method based on Fuzzy Clustering, which takes into consideration the semantic distance of terms transposing knowledge into a multidimensional space. GDARIM is also the data source input for an Opinion Mining system that needs preprocessed information from preselected sources.
Uploads
Papers by Daniela López De Luise
focuses on a prototype implementing a consciousness model for autonomous
mobile robots. The prototype is named MARCOS (Movement of Autonomous
Robotics Codelet System). Here is an in-depth analysis of this secondary controller
that complements the main close-loop controller providing complex information
about the current robot status, and suggesting alternative actions. This paper
describes the MARCOS prototype and its conscious subsystem, as well as statistics
on how Concepts can model real-world indoor environments.
protocol, a global and a case-based evaluation methodology are presented. There is also a kinetic pattern modeling in relation to
sound stimuli precisely defined. Findings here indicate that the proposed protocol is good for proper data collection required to
biomechanical and behavioral processing. It also enable to automate certain subtle information usually collected just by expert
evaluation. Thereby it is possible to evaluate relevant performance in patients with an automatic system.
Multimedia may provide a simple, short and effective way to explain them.
This paper aims to describe a virtual learning environment named MIDA. It
has been designed to train and evaluate a set of tips to improve teaching in
that area. LEARNITRON is an offspring of the project. It is a sub project for
automatic modelling learning profile of students according to the
combination of students profile, topics to be taught and current set of
multimedia tools. The goal of this paper is to sketch LEARNITRON and
how to improve multimedia tools for teaching, based on basic machine
learning techniques. The resulting model may be useful to define specific
features for better educational videos, and to acknowledge the right scope of
multimedia as a way to teach or train students in Engineering subjects based
not on personal opinion but on statistical findings
and persons. A good model requires the ability to adapt itself to several environ
mental variations and to preserve essential information about the area under scope.
This paper aims to present a proposal based on a Machine Learning extension for
timing named Harmonic Systems. A global description of the problem, its rele
vance and status of the field are also included
Many of them don't relay on vocal abilities because of the wide spectrum of
symptoms. But taking data from ML perspective may introduce new sources
of information for a better interpretation of patient's behaviour and helps
build new parameters to model individual performance. This paper presents
some initial findings in that direction based on the automatic evaluation of
real cases as a way to build a parameterised description of patients.
Preliminary analysis in patients suggest that the individual performance may
be compared with the universe of patients, it is possible to elaborate a
performance profiling and to model evolution profiles that are present in a
predefined universe of patients. The team is currently defining and testing
new audio and video protocols to collect additional parameters. From these
findings, it is possible to build a parametric reasoning model using MLW.
in their temporality, mainly to be able to make predictions. Interesting information arises when what is being examined is
not the complex information produced along the development of a process, but the temporality itself for the process. Such an
alternative point of view enables the relation of this analysis to a real process, and its application in the very moment the
events are taking place, knowing that there is a starting point of certain a priori knowledge.
This work is shows this alternative stance, from the temporal perspective. There are applications of concepts and
analogies brought from a different environment where time is handled in a more natural way. We shall call this approach
"Harmonic Systems". We shall present the basis for harmonic systems, the focused approach we propose as a type of time
mining, and its application
based on wavelets for image processing and an artificial neural network for data classification.
Handwritten Signature Validation (HSV) is a proven and accepted procedure for identity
validation by means of non-invasive and non-intrusive procedures (features which do biometrics
controversial as an authentication method). The experimental prototype is built on software
developed by AI Group Research Laboratories at Universidad de Palermo, project named PI-1
for handwritten signature recognition. The aim of the model is to improve automatic HSV
reliability through digital imaging algorithms and computational intelligence techniques. In the
next sections, a brief presentation of the work.
implementing a consciousness model for autonomous mobile robots. The prototype is named FIC (Fluent
Interactive Codelets framework). Here is an in-depth analysis of this secondary controller that complements
the main close-loop controller providing complex information about the current robot status, and suggesting
alternative actions. This paper describes the FIC prototype and its conscious subsystem, as well as statistics
on how Concepts can model real-world indoor environments.Autonomous mobile robots require efficient control of their movement. This paper focuses on a prototype
implementing a consciousness model for autonomous mobile robots. The prototype is named FIC (Fluent
Interactive Codelets framework). Here is an in-depth analysis of this secondary controller that complements
the main close-loop controller providing complex information about the current robot status, and suggesting
alternative actions. This paper describes the FIC prototype and its conscious subsystem, as well as statistics
on how Concepts can model real-world indoor environments.
method of the performance of autonomous mobile robots using different control algorithms. There is a first
description of the mobile robot problem and the importance of a standardized process for evaluation of robot
performances. There is a comparison between a simple navigation controller and an intelligent prototype based
on consciousness. The architecture main features are also outlined. Test cas
that has not replicated on dynamic areas of application. This paper presents the proposal of a research
project that identifies the main hurdles of this issue and present feasible solutions for the matter.
distributed systems. The work in progress presented here focuses on the representation of a distributed memory
for multi-robot systems. It states the main characteristics of a shared environment and provides suitable
interfaces to allow autonomous mobile robots to query, publish and exchange relevant information without
requiring a central data storage. The memory is based on a combination of fuzzy systems, distributed Self-
Organizing Maps (SOM), a new specific design information handshake between robots and a new model for
approach for world-map global administration.
and several statistics showing how it represents information extracted from Spanish dialogs. In order to get knowledge
quickly, dialogs are processed with a batch interface. The prototype has complex architecture with many components. It
is organized in several memories; each one models Spanish sentences that implement Morphosintactic Linguistic
Wavelets (MLW). Using a knowledge-representation mechanism, WIH can model answers automatically taking previous
dialogs with or between humans. Along with test cases description are also statistical analyses of the results obtained with
real dialogs. It is shown here that WIH prototype handles structures called E ce that are the kernel of its learning in a
human-like fashion, and that it is able to use it as a model for answers in future dialogs.
Support System (DDS) to be applied within transportation and
infrastructure enterprise level. The prototype retrieves from
different sources a set of parameters relative to the sector,
associated with micro and macro economic factors which directly
affect the activity, complemented with other studies related to
political, socio-cultural, technological and local regulatory
parameters. The architecture is centered on an Expert System
which provides the appropriate data mining tools set to be
applied for forecasting potential scenarios, while allowing the
configuration of the Meta-Expert Inferential System. The main
findings of the first system is intended to provide the second the
context and expertise of a data mining Specialist, while the
second system has the specific expertise allowing to advise the key
decision makers with the appropriate information based on the
risks to be faced in a particular business. The originality of this
prototype lies in the variables definition ever used in the sector
and the configuration of the Meta-Expert System.
La arquitectura se centra en un Sistema Experto que establece el conjunto apropiado de herramientas de Minería de Datos a aplicar para una correcta predicción de escenarios posibles, y configura el sistema inferencial de un Meta-Sistema Experto, al cual provee también de un conjunto de datos específicos para aplicar su conocimiento experto. Los eventuales hallazgos del primer sistema tienen como objetivo ofrecer al segundo sistema el contexto y expertise de un especialista en minería de datos, mientras que el segundo posee el conocimiento experto específico que le permite asesorar a los responsables de la toma de decisión la información apropiada para la toma de decisiones en base a los riesgos que deben afronta en un determinado negocio.
poco más que una tecnología de guerra que
buscaba reubicarse en la plaza científica y
tecnológica. El furioso éxito de los primeros
Sistemas Expertos (sistemas informáticos que
imitan a un experto humano en un área
específica) y algunas variantes
de Redes
Neuronales artificiales (especialmente aquellas
que permitían reconocimientos de patrones)
durante los años 60 culminó en un colapso casi
total luego de que la comunidad verificara en los
80 que aquellas tecnologías distaban mucho de
satisfacer las exigentes expectativas de ese
entonces.
for image processing and an Artificial Neural Network for data classification. Handwritten Signature
Validation (HSV) is a proven and accepted procedure for identity validation by means of non-invasive and non-
intrusive procedures. The experimental prototype is built on software developed by AI Group Research
Laboratories at Universidad de Palermo, project named PI-1 for handwritten signature validation. The aim of
the research is to improve automatic HSV reliability through digital imaging algorithms and computational
intelligence techniques. In the next sections, a brief presentation of the study, essays characteristics, results and
a solution design.
corpus efficiently is a matter of discussion and research in various fields. Search engines are in charge of grouping and
classifying web pages in terms of keywords. Web Crawler processes perform such tasks. During exploration the websites
are indexed according to different metrics, feeding the knowledge base of the search engine. Exploring hypertext and its
links presents an exponential order of growth in terms of computational time. This paper presents the GDARIM project a
Web Crawler that presents techniques that improve the accuracy of the information retrieved on a crawling process for a
given set of keywords. GDARIM implements an exploration method based on Fuzzy Clustering, which takes into
consideration the semantic distance of terms transposing knowledge into a multidimensional space. GDARIM is also the
data source input for an Opinion Mining system that needs preprocessed information from preselected sources.
focuses on a prototype implementing a consciousness model for autonomous
mobile robots. The prototype is named MARCOS (Movement of Autonomous
Robotics Codelet System). Here is an in-depth analysis of this secondary controller
that complements the main close-loop controller providing complex information
about the current robot status, and suggesting alternative actions. This paper
describes the MARCOS prototype and its conscious subsystem, as well as statistics
on how Concepts can model real-world indoor environments.
protocol, a global and a case-based evaluation methodology are presented. There is also a kinetic pattern modeling in relation to
sound stimuli precisely defined. Findings here indicate that the proposed protocol is good for proper data collection required to
biomechanical and behavioral processing. It also enable to automate certain subtle information usually collected just by expert
evaluation. Thereby it is possible to evaluate relevant performance in patients with an automatic system.
Multimedia may provide a simple, short and effective way to explain them.
This paper aims to describe a virtual learning environment named MIDA. It
has been designed to train and evaluate a set of tips to improve teaching in
that area. LEARNITRON is an offspring of the project. It is a sub project for
automatic modelling learning profile of students according to the
combination of students profile, topics to be taught and current set of
multimedia tools. The goal of this paper is to sketch LEARNITRON and
how to improve multimedia tools for teaching, based on basic machine
learning techniques. The resulting model may be useful to define specific
features for better educational videos, and to acknowledge the right scope of
multimedia as a way to teach or train students in Engineering subjects based
not on personal opinion but on statistical findings
and persons. A good model requires the ability to adapt itself to several environ
mental variations and to preserve essential information about the area under scope.
This paper aims to present a proposal based on a Machine Learning extension for
timing named Harmonic Systems. A global description of the problem, its rele
vance and status of the field are also included
Many of them don't relay on vocal abilities because of the wide spectrum of
symptoms. But taking data from ML perspective may introduce new sources
of information for a better interpretation of patient's behaviour and helps
build new parameters to model individual performance. This paper presents
some initial findings in that direction based on the automatic evaluation of
real cases as a way to build a parameterised description of patients.
Preliminary analysis in patients suggest that the individual performance may
be compared with the universe of patients, it is possible to elaborate a
performance profiling and to model evolution profiles that are present in a
predefined universe of patients. The team is currently defining and testing
new audio and video protocols to collect additional parameters. From these
findings, it is possible to build a parametric reasoning model using MLW.
in their temporality, mainly to be able to make predictions. Interesting information arises when what is being examined is
not the complex information produced along the development of a process, but the temporality itself for the process. Such an
alternative point of view enables the relation of this analysis to a real process, and its application in the very moment the
events are taking place, knowing that there is a starting point of certain a priori knowledge.
This work is shows this alternative stance, from the temporal perspective. There are applications of concepts and
analogies brought from a different environment where time is handled in a more natural way. We shall call this approach
"Harmonic Systems". We shall present the basis for harmonic systems, the focused approach we propose as a type of time
mining, and its application
based on wavelets for image processing and an artificial neural network for data classification.
Handwritten Signature Validation (HSV) is a proven and accepted procedure for identity
validation by means of non-invasive and non-intrusive procedures (features which do biometrics
controversial as an authentication method). The experimental prototype is built on software
developed by AI Group Research Laboratories at Universidad de Palermo, project named PI-1
for handwritten signature recognition. The aim of the model is to improve automatic HSV
reliability through digital imaging algorithms and computational intelligence techniques. In the
next sections, a brief presentation of the work.
implementing a consciousness model for autonomous mobile robots. The prototype is named FIC (Fluent
Interactive Codelets framework). Here is an in-depth analysis of this secondary controller that complements
the main close-loop controller providing complex information about the current robot status, and suggesting
alternative actions. This paper describes the FIC prototype and its conscious subsystem, as well as statistics
on how Concepts can model real-world indoor environments.Autonomous mobile robots require efficient control of their movement. This paper focuses on a prototype
implementing a consciousness model for autonomous mobile robots. The prototype is named FIC (Fluent
Interactive Codelets framework). Here is an in-depth analysis of this secondary controller that complements
the main close-loop controller providing complex information about the current robot status, and suggesting
alternative actions. This paper describes the FIC prototype and its conscious subsystem, as well as statistics
on how Concepts can model real-world indoor environments.
method of the performance of autonomous mobile robots using different control algorithms. There is a first
description of the mobile robot problem and the importance of a standardized process for evaluation of robot
performances. There is a comparison between a simple navigation controller and an intelligent prototype based
on consciousness. The architecture main features are also outlined. Test cas
that has not replicated on dynamic areas of application. This paper presents the proposal of a research
project that identifies the main hurdles of this issue and present feasible solutions for the matter.
distributed systems. The work in progress presented here focuses on the representation of a distributed memory
for multi-robot systems. It states the main characteristics of a shared environment and provides suitable
interfaces to allow autonomous mobile robots to query, publish and exchange relevant information without
requiring a central data storage. The memory is based on a combination of fuzzy systems, distributed Self-
Organizing Maps (SOM), a new specific design information handshake between robots and a new model for
approach for world-map global administration.
and several statistics showing how it represents information extracted from Spanish dialogs. In order to get knowledge
quickly, dialogs are processed with a batch interface. The prototype has complex architecture with many components. It
is organized in several memories; each one models Spanish sentences that implement Morphosintactic Linguistic
Wavelets (MLW). Using a knowledge-representation mechanism, WIH can model answers automatically taking previous
dialogs with or between humans. Along with test cases description are also statistical analyses of the results obtained with
real dialogs. It is shown here that WIH prototype handles structures called E ce that are the kernel of its learning in a
human-like fashion, and that it is able to use it as a model for answers in future dialogs.
Support System (DDS) to be applied within transportation and
infrastructure enterprise level. The prototype retrieves from
different sources a set of parameters relative to the sector,
associated with micro and macro economic factors which directly
affect the activity, complemented with other studies related to
political, socio-cultural, technological and local regulatory
parameters. The architecture is centered on an Expert System
which provides the appropriate data mining tools set to be
applied for forecasting potential scenarios, while allowing the
configuration of the Meta-Expert Inferential System. The main
findings of the first system is intended to provide the second the
context and expertise of a data mining Specialist, while the
second system has the specific expertise allowing to advise the key
decision makers with the appropriate information based on the
risks to be faced in a particular business. The originality of this
prototype lies in the variables definition ever used in the sector
and the configuration of the Meta-Expert System.
La arquitectura se centra en un Sistema Experto que establece el conjunto apropiado de herramientas de Minería de Datos a aplicar para una correcta predicción de escenarios posibles, y configura el sistema inferencial de un Meta-Sistema Experto, al cual provee también de un conjunto de datos específicos para aplicar su conocimiento experto. Los eventuales hallazgos del primer sistema tienen como objetivo ofrecer al segundo sistema el contexto y expertise de un especialista en minería de datos, mientras que el segundo posee el conocimiento experto específico que le permite asesorar a los responsables de la toma de decisión la información apropiada para la toma de decisiones en base a los riesgos que deben afronta en un determinado negocio.
poco más que una tecnología de guerra que
buscaba reubicarse en la plaza científica y
tecnológica. El furioso éxito de los primeros
Sistemas Expertos (sistemas informáticos que
imitan a un experto humano en un área
específica) y algunas variantes
de Redes
Neuronales artificiales (especialmente aquellas
que permitían reconocimientos de patrones)
durante los años 60 culminó en un colapso casi
total luego de que la comunidad verificara en los
80 que aquellas tecnologías distaban mucho de
satisfacer las exigentes expectativas de ese
entonces.
for image processing and an Artificial Neural Network for data classification. Handwritten Signature
Validation (HSV) is a proven and accepted procedure for identity validation by means of non-invasive and non-
intrusive procedures. The experimental prototype is built on software developed by AI Group Research
Laboratories at Universidad de Palermo, project named PI-1 for handwritten signature validation. The aim of
the research is to improve automatic HSV reliability through digital imaging algorithms and computational
intelligence techniques. In the next sections, a brief presentation of the study, essays characteristics, results and
a solution design.
corpus efficiently is a matter of discussion and research in various fields. Search engines are in charge of grouping and
classifying web pages in terms of keywords. Web Crawler processes perform such tasks. During exploration the websites
are indexed according to different metrics, feeding the knowledge base of the search engine. Exploring hypertext and its
links presents an exponential order of growth in terms of computational time. This paper presents the GDARIM project a
Web Crawler that presents techniques that improve the accuracy of the information retrieved on a crawling process for a
given set of keywords. GDARIM implements an exploration method based on Fuzzy Clustering, which takes into
consideration the semantic distance of terms transposing knowledge into a multidimensional space. GDARIM is also the
data source input for an Opinion Mining system that needs preprocessed information from preselected sources.