IJCAI-05

19th INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE



Multi-Agent Information Retrieval and Recommender Systems

IJCAI-2005 Workshop

[Overview] [Topics] [Schedule] [Submission] [Deadlines]
[Organizing committee] [Program committee] [Contact]

Workshop Schedule

Accepted papers

Overview

Providing effective access to on-line information is a problem that brings together many Artificial Intelligence topics: knowledge representation, natural language, user modelling and machine learning, as well as distributed computing, Human-Computer Interaction and databases. Over the last few years, artificial intelligence techniques have provided substantial progress toward solving the problems of information access through the following three domains: models and techniques for multi-agents systems, information retrieval and recommender systems.

In the first domain, there is a large community that develops architectures, language models and techniques for interaction and communication between agents. An important focus for the information retrieval domain has been the development of techniques for coping with the heterogeneous and dynamic information space represented by the World-Wide Web and the impact of new retrieval devices and interfaces. However, information retrieval systems do not traditionally care about the individual searcher, preferring instead to focus on the development of global retrieval techniques rather than those adapted for the needs of the individual. Consequently, recommender systems research has focused on the interaction between information retrieval and user modelling in order to provide a more personalized and proactive retrieval experience and to help users choose between retrieval alternatives and to refine their queries.

In the future, one can easily imagine a virtual organization of agents with specific tasks such as profile acquisition, web searching and recommendation making. These agents cooperate and negotiate in order to satisfy their users while they put their precious human cycles to a better use. The agents interact to complement their partial solutions or to solve conflicts that may arise. The integration of agent technology will help create and maintain this virtual organization.

Rapidly evolving computer technology, coupled with the exponential growth of the services and information available on the Internet, have already brought us to the point where hundreds of millions of people should have fast, pervasive access to phenomenal amounts of information. The challenge of complex environments is therefore obvious: software is expected to do more in more situations, there is a variety of users, there is a variety of interactions and there is a variety of resources and goals. To cope with such environments, the promise of Multi-Agent Systems is becoming highly attractive.

This workshop will present an opportunity for researchers working in these three domains to get together and share recent developments and techniques in order to identify the critical problems and the most promising research avenues. In particular the workshop will emphasize the interaction between these different research areas, with a view to establishing a framework for more flexible multi-agent retrieval and recommendation solutions.

[Back to top]

Topics

The following issues may be addressed: Special consideration will be given to contributions aimed at incorporating multiple models and techniques. We are particularly interested in multi-agent systems dedicated to information retrieval and recommender systems.

Application areas include, but are not limited to:

[Back to top]

Schedule

TimeEvent (David Hume Tower Conference Room)
 
9:00-9:15 Welcome
 
9:15-9:40 David McSherry
Conversational CBR in multi-agent recommendation
9:40-10:05 Tong Zheng and Randy Goebel
An NCM-based framework for navigation recommender systems
10:05-10:30 Conor Hayes, Paolo Avesani and Marco Cova
Language games: Learning shared concepts among distributed information agents
 
10:30-11:00 Coffee Break
 
11:00-11:25 Fabiana Lorenzi, Daniela Scherer dos Santos and Ana L. C. Bazzan
Negotiation for task allocation among agents in case-base recommender systems: A swarm-intelligence approach
11:25-11:50 Wolfgang Woerndl and Georg Groh
A proposal for an agent-based architecture for context-aware personalization in the Semantic Web
11:50-12:15 Alexander Birukov, Enrico Blanzieri and Paolo Giorgini
Implicit: A recommender system that uses implicit knowledge to produce suggestions
 
12:15-13:45 Lunch
 
13:45-14:10 Robin Burke and Bamshad Mobasher
Trust and bias in multi-agent recommender systems
14:10-14:35 Elhadi Shakshuki, André Trudel, Yiqing Xu and Boya Li
A probabilistic temporal interval algebra based multi-agent scheduling system
14:35-15:00 Maria Salamó, Barry Smyth, Kevin McCarthy, James Reilly and Lorraine McGinty
Reducing critiquing repetition in conversational recommendation
 
15:00-15:25 Coffee Break
 
15:25-15:50 Timothy A. Musgrove and Robin H. Walsh
Utilizing multiple agents in dynamic product recommendations
15:50-16:15 William H. Hsu
Relational graphical models for collaborative filtering and recommendation of computational workflow components
16:15-16:40 François Paradis, Jian-Yun Nie and Arman Tajarobi
Discovery of business opportunities on the Internet with information extraction
16:40-17:05 Joel Pinho Lucas, Beatriz Wilges and Ricardo Azambuja Silveira
Making use of FIPA multiagent architecture to develop animated pedagogical agents inside intelligent learning environments
17:05-17:30 Hugues Tremblay-Beaumont and Esma Aïmeur
Feature combination in a recommender system using distributed items: The case of JukeBlog
 
17:35-18:00 Wrap-up
[Back to top]

Submission Instructions

Prospective participants should submit a research paper or a brief position paper describing their interests in two or more of the workshop’s domains: multi-agents, information retrieval and recommender systems. Papers should be submitted electronically to aimeur@iro.umontreal.ca in Microsoft Word or Adobe PDF. Please refer to the guidelines on the IJCAI-05 web site. Full papers should range from five to eight pages, and position papers from two to five pages. Papers should be submitted electronically to aimeur@iro.umontreal.ca using "IJCAI-05 Workshop Submission" as the Subject of the email message.

Important Dates and Deadlines

[Back to top]

Organizing Committee

Program Committee

[Back to top]

Primary contact

For questions and suggestions, please contact Esma Aïmeur

Email: aimeur@iro.umontreal.ca

URL: http://www.iro.umontreal.ca/~aimeur

Phone: +1 (514) 343-6794

Fax: +1 (514) 343-5834

Université de Montréal
Département d'informatique et de recherche opérationnelle
Pavillon André-Aisenstadt
C.P. 6128, Succ. Centre-Ville, Montréal (QC)
H3C 3J7 Canada

[Back to top]