Hans-J. Lenz - Böcker

Ten years ago Bill Gale of AT&T Bell Laboratories was primary organizer of the first Workshop on Artificial Intelligence and Statistics. In the early days of the Workshop series it seemed clear that researchers in AI and statistics had common interests, though with different emphases, goals, and vocabularies. In learning and model selection, for example, a historical goal of AI to build autonomous agents probably contributed to a focus on parameter-free learning systems, which relied little on an external analyst's assumptions about the data. This seemed at odds with statistical strategy, which stemmed from a view that model selection methods were tools to augment, not replace, the abilities of a human analyst. Thus, statisticians have traditionally spent considerably more time exploiting prior information of the environment to model data and exploratory data analysis methods tailored to their assumptions. In statistics, special emphasis is placed on model checking, making extensive use of residual analysis, because all models are 'wrong', but some are better than others. It is increasingly recognized that AI researchers and/or AI programs can exploit the same kind of statistical strategies to good effect. Often AI researchers and statisticians emphasized different aspects of what in retrospect we might now regard as the same overriding tasks.

Planning of actions based on decision theory is a hot topic for many disciplines. Seemingly unlimited computing power, networking, integration and collaboration have meanwhile attracted the attention of fields like Machine Learning, Operations Research, Management Science and Computer Science. Software agents of e-commerce, mediators of Information Retrieval Systems and Database based Information Systems are typical new application areas. Until now, planning methods were successfully applied in production, logistics, marketing, finance, management, and used in robots, software agents etc. It is the special feature of the book that planning is embedded into decision theory, and this will give the interested reader new perspectives to follow-up.

The book aims to merge Computational Intelligence with Data Mining, which are both hot topics of current research and industrial development, Computational Intelligence, incorporates techniques like data fusion, uncertain reasoning, heuristic search, learning, and soft computing. Data Mining focuses on unscrambling unknown patterns or structures in very large data sets. Under the headline Discovering Structures in Large Databases the book starts with a unified view on Data Mining and Statistics - A System Point of View. Two special techniques follow: Subgroup Mining, and Data Mining with Possibilistic Graphical Models. Data Fusion and Possibilistic or Fuzzy Data Analysis is the next area of interest. An overview of possibilistic logic, nonmonotonic reasoning and data fusion is given, the coherence problem between data and non-linear fuzzy models is tackled, and outlier detection based on learning of fuzzy models is studied. In the domain of Classification and Decomposition adaptive clustering and visualisation of high dimensional data sets is introduced. Finally, in the section Learning and Data Fusion learning of special multi-agents of virtual soccer is considered.The last topic is on data fusion based on stochastic models.