Invited Talks

Pádraig Cunningham, Barry Smyth:
An Analysis of Research Themes in the CBR Conference Literature

Web page with the slides in pdf format

After fifteen years of CBR conferences, we examine the themes that have evolved in CBR research as revealed by the implicit and explicit relationships between the conference papers. We have examined a number of metrics for demonstrating connections between papers and between authors and have found that a clustering based on co-citation of papers appears to produce the most meaningful organisation. We have employed an Ensemble Non-negative Matrix Factorisation (NMF) approach that produces a "soft" hierarchical clustering, where papers can belong to more than one cluster. This is useful as papers can naturally relate to more than one research area. We have produced timelines for each of these clusters that highlight influential papers and illustrate the life-cycle of research themes over the last fifteen years. The insights afforded by this analysis will be presented in detail. In addition to the analysis of the sub-structure of CBR research, we will also present some global statistics on the CBR conference literature.

Pádraig Cunningham is Professor of Knowledge and Data Engineering in the School of Computer Science and Informatics at University College Dublin. He has a B.E. and M.Eng.Sci. from NUI Galway and a Ph.D. from Dublin University which he received in 1989. He became a Fellow of ECCAI (European Coordinating Committee on Artificial Intelligence) in 2004. After completing his PhD he worked with Digital Equipment Corporation as a Software Engineer and with Hitachi Europe Ltd. as a Research Scientist. He joined Trinity College Dublin as a lecturer in 1992 and moved to the School of Computer Science and Informatics in University College Dublin in 2006. He is a former director of start-up companies New World Commerce and Prediction Dynamics. Dr. Cunningham has published over 140 peer-reviewed papers in the general area of applied AI, focusing on machine learning and case-based reasoning systems for decision support in engineering, e-commerce, finance and medicine.

Barry Smyth holds the Digital Chair of Computer Science in the School of Computer Science and Informatics. Barry's research interests cover many aspects of artificial intelligence and information retrieval, including case-based reasoning, recommender systems, user modeling, and personalized user interfaces. He has published more than 300 peer-reviewed articles and has received a number of awards for his research. In 1999 Barry co-founded ChangingWorlds Ltd. to bring personalization technologies to mobile operators. He continues to serve as the company's Chief Scientist. ChangingWorlds now employs 150 people in offices around the world, and counts 50 of the worlds leading operators as among its clients. Barry has recently helped to secure in excess of €16m in funding for CLARITY, the Centre for Sensor Web Technologies, where he will serve as Director. CLARITY will explore the combination of next-generation adaptive sensing and information discovery technologies in a range of application areas, including environmental monitoring, personalized healthcare, and social media.

Enric Plaza:
Semantics and Experience in the Future Web

Web page with the slides in pdf format

The Web is a vibrant environment for innovation in computer science, AI, and social interaction; these innovations come in such great number and speed that it is unlikely to follow them. This talk will focus on some emerging aspects on the web that are an opportunity and challenge for Case-based Reasoning, specifically the large amount of "experiences" that individual people share in the Web. The talk will try to characterize this "experiences", these bits of practical knowledge that go from simple but practical facts to complex problem solving descriptions. Then, I'll focus on how CBR ideas could be brought to bear in sharing and reusing this experiential knowledge, and finally on the challenging issues that have to be addressed for that purpose.

Enric Plaza Cervera holds a Ph.D. in Computer Science by the Technical University of Catalonia (UPC) and is Research Professor of the Spanish Council for Scientific Research at the Barcelona Artificial Intelligence Research Institute (IIIA) since 1988. He has worked on knowledge acquisition, case-based reasoning (CBR), machine learning and multiagent systems in a dozen of European and Spanish projects. He has chaired three international conferences on A.I. fields and has authored over 100 scientific papers. His research is now focused on case-based reasoning and argumentation in multiagent system, reuse techniques for CBR, and social web mining. He is member of a dozen program committees annually for international conferences, plus other conferences and workshops. He is an ECCAI Fellow and has served as Chairman of the Board of Trustees of the ACIA (Catalan Association for Artificial Intelligence) during four years and frequently publishes articles for AI popularization.

Isabelle Bichindaritz:
CBR in the Health Sciences: Why it Matters for the Health Sciences and for CBR

.zip archive with the slides in pdf format (2.5 MB)

Biomedical domains have been an application domain of choice for artificial intelligence (AI) since its pioneering years in expert systems. One simple explanation to this phenomenon is the dominant industry market share of healthcare. Following in AI’s tracks, case-based reasoning (CBR) has been abundantly applied to the health sciences domain and has produced an excellent as well as varied set of publications, which has fostered CBR research innovation to answer some of the research issues associated with this complex domain. Some notable examples are synergies with other AI methodologies, and in particular with ontologies and with data mining, the study of the temporal dimension in CBR, the processing of multimedia cases, and novel tasks for CBR such as parameter setting. However CBR has a major challenge to take on in the health sciences: how to position itself with regard to statistics for processing data? Some claim that CBR proposes an alternative viewpoint on the concept of evidence in biomedicine. Others that CBR and statistics complement one another in this domain. In any case, an interesting question to study is whether CBR could become one day as fundamental to the health sciences as statistics is today? This question in particular broadens the health sciences challenges to a universal scope.

Isabelle Bichindaritz is an assistant professor at the University of Washington, Institute of Technology, Computing and Software Systems, in Tacoma, WA, USA since 2002, where she directs the Laboratory of Informatics and Artificial Intelligence. She holds a Ph.D. in Computer Science from the Université René Descarte- Paris V (1994). Her research focuses on intelligent learning systems in biology and medicine, with a particular emphasis on case-based reasoning and data mining as well as on biomedical applications of artificial intelligence. She has co-organized the last five workshops on Case-based Reasoning in the Life Sciences, co-located with the International Conferences on Case-based Reasoning and the European Conferences on Case-based Reasoning, and organized the first two workshops on Data Mining in the Life Sciences, co-located with the Industrial Conferences on Data Mining. Dr. Bichindaritz has published over 90 scientific papers, and has edited or co-edited several journals special issues, such as an Artificial Intelligence in Medicine journal special issue, a Computational Intelligence special issue, and an Applied Intelligence special issue on Case-based Reasoning in the Life Sciences.