With the advent of new technologies and business practices, including the explosion of electronic commerce and the trend toward mass personalization, businesses are searching for new ways to make sense of massive, and increasing, amounts of data. Data mining allows businesses to ascertain patterns in data that indicate consumer buying habits, competitor strategies, credit worthiness, incidence of fraud, and other information of strategic importance. This mini-track seeks to provide a forum for researchers to share investigations and ideas in the diverse field of data mining. Current research in data mining explores areas related to algorithm development and performance, the data mining process, and the impact of data mining on business practices, among others. Information Retrieval (IR) algorithms support computerized search of large document collections (i.e., millions of documents) to retrieve small subsets relevant to the user’s information needs. Examples are book searching in electronic bookstores and auctions, digital library catalogues and Internet search engines. IR Application areas include: cross language retrieval, speech/broadcast retrieval, text categorization, and text summarization. IR algorithms are subject to objective testing and evaluation for hundreds of queries on millions of documents (the TREC set and conferences, for example).

Possible Topics may include, but are not limited to, the following areas:

• Mining of multimedia data, including image, video and sound data
• New/hybrid algorithms for data mining
• Performance evaluation of data mining methods
• Aligning data mining with task objectives and business strategy
• The impact of data mining on e-commerce strategies
• Mining of internet-based data
• Domain- and problem-specific data mining
• Data mining applications in industry
• Exploring the data mining process, including issues of data selection, cleaning and preprocessing, and interpretation of results
• Information Retrieval algorithms for electronic commerce and other applications
• Classifying the WWW with text categorization and clustering
• Analyzing WWW query logs for persistent patterns and query clusters
• Non-English language and cross-language information retrieval
• Text summarization and gisting

William E. Spangler (*)
West Virginia University
(304) 293-7933
wspangle@wvu.edu

H. Michael Chung
California State University, Long Beach
(562)985-7691
hmchung@csulb.edu

Fredric C. Gey
University of California, Berkeley
(510) 642-6571
gey@ucdata.berkeley.edu