Donald Kraft

Abstract: We present some results from research in applying genetic programming to determine an optimal query for classifying text documents into two subsets of documents deemed relevant to a user query and not relevant to that query based on relevance feedback. We shall first briefly introduce the concept of textual information retrieval (IR), look at computing issues related to IR, and note some interesting applications of IR to other problems. We note some key issues needed for proper retrieval, such as keywords, relevance, relevance feedback, and performance measures for IR. We then turn to a model of fuzzy Boolean retrieval and show how genetic programming can be used to find an optimal Boolean query to classify documents as relevant or nonrelevant. We conclude with some ideas for current research, such as fuzzy rough sets and fuzzy clustering and rule generation, and some indications of areas for future research.

Professor Donald H. Kraft holds the BSIE, MSIE, and Ph.D. degrees from Purdue University in Industrial Engineering with a specialization in operations research. He is currently a professor in the Department of Computer Science, an adjunct professor of library and information science, and a former department chair in computer science at Louisiana State University (LSU). He is also currently a Distinguished Visiting Professor at the U.S. Air Force Academy. He has been a faculty member at the University of Maryland and a visiting faculty member at the University of California - Berkeley, Indiana University, and the University of California - Los Angeles.

Professor Kraft's primary research interests lie in information retrieval models, and the application of artificial intelligence tools, such as fuzzy sets, rough sets, and genetic programming, to such retrieval models. He has co-authored three books, including one on information retrieval, several book chapters, scholarly journal articles, and several refereed conference proceedings, and has given many other presentations and talks. He has been awarded the American Society for Information Science and Technology (ASIST) Best Information Science Book Award and the ASIST Research Award; and he is both an IEEE and an AAAS Fellow. In addition, he has recently been given an LSU Distinguished Faculty award.

Professor Kraft is the Editor of the Journal of the American Society for Information Science and Technology (JASIST), for which he won the ASIST Watson Davis Award for service to the Society. He is also a past president of ASIST. In addition, he is an Associate Editor of the IEEE Transactions on Fuzzy Systems. He has recently chaired conferences for NAFIPS and ACM/SIGIR. He is an Editorial Board Member for Information Retrieval, an Associate Editor of Scientometrics, and is on the Editorial Review Board of International Journal of Operations and Quantitative Management (IJOQM). He was awarded a LSU Community Volunteer Award for volunteer activities on and off campus, including church and United Way work.

Professor Kraft currently teaches courses in software systems development, programming languages, numerical methods, and information retrieval.