Back to first pageBack to first page Centre for Artificial Intelligence of UNL
Browse our site
You are here:

Publication details

Go back
Publication details
Main information
Evolution Prospection with Intention Recognition via Computational Logic
June 2009
han-msc-thesis09
The thesis concerns the problem of modelling evolving prospective agent systems. A prospective agent looking ahead into the future is confronted with having several courses of evolution, and needs to prefer amongst them to determine the best to follow. Based on historical information and quantitative and qualitative a posteriori evaluation of possible evolutions, the agent is equipped with evolution-level preferences mechanism. To enable the agent to evolve, we provide modelling its evolving knowledge base, and environment triggering of active goals, context-sensitive preferences and integrity constraints. To allow an evolving agent acting under uncertainty, P-log is employed to represent probabilistic knowledge. Agents are enhanced with intention recognition, via combination of Causal Bayes Networks and plan attribution. Examples illustrate concepts and features. We show how the agent system can be applied to model morality and provide support for elderly people.
M. Sc. dissertation
Han The Anh
Luís Moniz Pereira
Technical University of Dresden, Germany
-
-
Export formats
Han The Anh, Evolution Prospection with Intention Recognition via Computational Logic, Luís Moniz Pereira (superv.), Technical University of Dresden, Germany, June 2009.
<b><a href="/people/members/view.php?code=cdc7090d1f84f56c0671baa36e87bd77" class="author">Han The Anh</a></b>, <u>Evolution Prospection with Intention Recognition via Computational Logic</u>, <a href="/people/members/view.php?code=6175f826202ff877fba2ad77784cb9cb" class="supervisor">Luís Moniz Pereira</a> (superv.), Technical University of Dresden, Germany, June 2009.
@mastersthesis {han-msc-thesis09, author = {Han The Anh}, title = {Evolution Prospection with Intention Recognition via Computational Logic}, school = {Technical University of Dresden, Germany}, note = {Lu\'{\i}s Moniz Pereira (superv.); }, abstract = {The thesis concerns the problem of modelling evolving prospective agent systems. A prospective agent looking ahead into the future is confronted with having several courses of evolution, and needs to prefer amongst them to determine the best to follow. Based on historical information and quantitative and qualitative a posteriori evaluation of possible evolutions, the agent is equipped with evolution-level preferences mechanism. To enable the agent to evolve, we provide modelling its evolving knowledge base, and environment triggering of active goals, context-sensitive preferences and integrity constraints. To allow an evolving agent acting under uncertainty, P-log is employed to represent probabilistic knowledge. Agents are enhanced with intention recognition, via combination of Causal Bayes Networks and plan attribution. Examples illustrate concepts and features. We show how the agent system can be applied to model morality and provide support for elderly people.}, keywords = {Agents, Evolution Prospection, Elder Care, Intention Recognition}, month = {June}, year = {2009}, }
Publication's urls
/publications/view.php?code=e3a9f351b5c622fcb559951f40f4573f
/publications/view.php?code=han-msc-thesis09
Go back
Centre for Artificial Intelligence of UNL
Departamento de Informática, FCT/UNL
Quinta da Torre 2829-516 CAPARICA - Portugal
Tel. (+351) 21 294 8536 FAX (+351) 21 294 8541

Fundacao_FCT