Powered by OpenAIRE graph
Found an issue? Give us feedback
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/ Recolector de Cienci...arrow_drop_down
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/
Recolector de Ciencia Abierta, RECOLECTA
Doctoral thesis . 2010 . Peer-reviewed
License: CC BY NC ND
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/
Recolector de Ciencia Abierta, RECOLECTA
Doctoral thesis . 2010
License: CC BY NC ND
versions View all 2 versions
addClaim

Modelado automático del comportamiento de agentes inteligentes

Authors: Iglesias Martínez, José Antonio;

Modelado automático del comportamiento de agentes inteligentes

Abstract

Las teorías más recientes sobre el cerebro humano confirman que un alto porcentaje de su capacidad es utilizado para predecir el futuro, incluyendo el comportamiento de otras personas. Para actuar de la forma más adecuada en un contexto social, los humanos tratan de reconocer el comportamiento de las personas que les rodean y así hacer predicciones basadas en estos reconocimientos. Cuando este proceso se lleva a cabo por agentes software, se conoce como modelado de agentes donde un agente puede ser un robot, un agente software o un humano. El modelado de agentes es un proceso que permite a un agente extraer y representar conocimiento (comportamiento, creencias, metas, acciones, planes, etcétera) de otros agentes en un entorno determinado. Un agente capaz de reconocer el comportamiento de otros, puede realizar diversas tareas como predecir el comportamiento futuro de los agentes observados, coordinarse con ellos, facilitarles la ejecución de sus acciones o detectar sus posibles errores. Si este reconocimiento puede ser realizado de forma automática, su utilidad puede ser muy relevante en muchos dominios. En esta tesis doctoral se aborda la tarea de adquirir automáticamente conocimiento acerca del comportamiento de otros agentes inteligentes. Actualmente, las técnicas para modelar el comportamiento de otros agentes están comenzando a surgir de forma importante en el campo de la Inteligencia Artificial. Cabe destacar que en la mayoría de las investigaciones actuales, se proponen modelados no generales de un determinado tipo de agentes en un dominio concreto, es decir, modelados ad hoc. Esta tesis doctoral presenta tres enfoques diferentes para el modelado automático del comportamiento de agentes inteligentes basado en la identificación de patrones en un comportamiento observado. Estos enfoques permitirán que un agente situado en un entorno determinado, sea capaz de adquirir conocimiento acerca de otros agentes situados en el mismo entorno. Cada enfoque propuesto posee características particulares que le permiten adecuarse a un tipo de dominio, lo que implica que se puede adquirir conocimiento de otros agentes en diversos Sistema Multiagente. Los tres enfoques propuestos transforman las observaciones del comportamiento de uno o varios agentes en una secuencia de eventos que lo definen. Esta secuencia es analizada con la finalidad de obtener su correspondiente modelo de comportamiento. De esta forma, en esta tesis doctoral, la tarea de modelado e identificación del comportamiento de uno o varios agentes es tratado, principalmente, como un problema de minería de secuencias de eventos. La aplicación de cada enfoque propuesto en dominios muy diferentes demuestra su generalidad.----------------------------------------------------------------------------------- There are new theories which claim that a high percentage of the human brain capacity is used for predicting the future, including the behavior of other humans. Planning for future needs, not just current ones, is one of the most formidable human cognitive achievements. To make good decisions in a social context, humans often need to recognize the plan underlying the behavior of others, and make predictions based on this recognition. This process, when carried out by software agents, is known as agent modeling where an agent can be a software agent, a robot or a human being. Agent modeling is the process of extracting and representing knowledge (behavior, beliefs, goals, actions, plans, etcetera) from other agents. By recognizing the behavior of others, many different tasks can be performed, such as to predict their future behavior, to coordinate with them or to assist them. This behavior recognition can be very useful in many applications if it can be done automatically. This thesis is framed in the field of agent behavior modeling. Most existing techniques for plan recognition assume the availability of carefully hand-crafted plan libraries, which encode the a-priori known behavioral repertoire of the observed agents; during run-time, plan recognition algorithms match the observed behavior of the agents against the plan-libraries, and matches are reported as hypotheses. Unfortunately, techniques for automatically acquiring plan-libraries from observations, e.g., by learning or data-mining, are only beginning to emerge. This thesis presents three different approaches for creating automatically the model of an agent behavior based on the analysis of its atomic behaviors. Each approach is suitable for different purposes, but in all of them, the observations of an agent behavior are transformed into a sequence of events which is analyzed in order to get the corresponding behavior model. Therefore, in this thesis, the problem of behavior classification is examined as a problem of learning to characterize the behavior of an agent in terms of sequences of atomic behaviors. In order to demonstrate the generalization of the proposed approaches, their performance has been experimentally evaluated in different domains.

Country
Spain
Related Organizations
Keywords

Informática, Modelado automático de agentes, Inteligencia artificial, Agentes inteligentes

  • BIP!
    Impact byBIP!
    selected citations
    These citations are derived from selected sources.
    This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    0
    popularity
    This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
    Average
    influence
    This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    Average
    impulse
    This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
    Average
    OpenAIRE UsageCounts
    Usage byUsageCounts
    visibility views 29
    download downloads 13
  • 29
    views
    13
    downloads
    Powered byOpenAIRE UsageCounts
Powered by OpenAIRE graph
Found an issue? Give us feedback
visibility
download
selected citations
These citations are derived from selected sources.
This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Citations provided by BIP!
popularity
This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
BIP!Popularity provided by BIP!
influence
This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Influence provided by BIP!
impulse
This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
BIP!Impulse provided by BIP!
views
OpenAIRE UsageCountsViews provided by UsageCounts
downloads
OpenAIRE UsageCountsDownloads provided by UsageCounts
0
Average
Average
Average
29
13
Green