M.J. Martín-bautista and M.A. Vila
Department of Computer Science and Artificial Intelligence, Granada University
Avda. Andalucía 38, Granada, 18071, Spain
mbautis@decsai.ugr.es
phone: +34-958-244019, fax: +34-958-243317

D.H. Kraft and J. Chen
Department of Computer Science
Louisiana State University
Baton Rouge, LA 70803-4020, USA

J. Cruz
Puleva Salud Inc.
Camino de Purchil 66, 18004
Granada, Spain

Abstract
We present a study of the role of user profiles and fuzzy logic in web retrieval processes. Flexibility for user interaction and for adaptation in profile construction becomes an important issue. We focus our study on user profiles, including creation, modification, storage, clustering and interpretation. We also consider the role of fuzzy logic and other soft computing techniques to improve user profiles. Extended profiles contain additional information related to the user that can be used to personalize and customize the retrieval process as well as the web site. Web mining processes can be carried out by means of fuzzy clustering of these extended profiles and fuzzy rule construction. Fuzzy inference can be used in order to modify queries and extract knowledge from profiles with marketing purposes within a web framework. An architecture of a portal that supports web mining technology is also presented.

Keywords: User profile, Web Mining, Fuzzy Logic, Fuzzy Clustering, Soft Computing, Information Retrieval

Extensions and Generalizations of Combining Fuzzy Sets and Rough Sets: Vocabulary Mining for Information Retrieval

Padmini Srinivasan	Donald H. Kraft
School of Library and Information Science	Department of Computer Science
The University of Iowa	Louisiana State University
Iowa City, IA 52242, USA	Baton Rouge, LA 70803-4020, USA

Abstract
The concept of vocabulary mining deals with using the domain vocabulary, perhaps a controlled vocabulary, with the goal of improving the performance of an information retrieval system in response to a user query. Previous research in this area has shown the feasibility, at least theoretically, of using rough sets to accomplish this mining. One can attempt to generalize, specialize, or perform other types of vocabulary-based transformations on the query to obtain a better query, better in terms of retrieval results. We extend our prior work in looking at generalized rough sets and coordinating multiple views of the vocabulary relationships. A discussion of applying this extended model to the Unified Medical Language System (UMLS) is also presented.

Technical Paper, Department of Computer Science, Louisiana State University, Baton Rouge, LA, USA, 2003

Jianhua Chen and Donald H. Kraft	Maria J. Martin-Bautista and Maria-Amparo Vila
Dept. of Computer Science	Dept. of Computer Science and AI
Louisiana State University	University of Granada
Baton Rouge, LA 70803-4020 USA	Granada 18071, SPAIN
{jianhua, kraft}@bit.csc.lsu.edu	{mbautista, vila}@decsai.ugr.es

Abstract
In this chapter, we present a unified framework that combines fuzzy rule induction and inference with textual information retrieval using user profiles. Fuzzy rules are extracted from the fuzzy clusters discovered by the fuzzy C-means clustering method. These rules can be used to characterize the semantic connections between keywords in a set of textual documents, and thus the rules can be used to improve the user queries for better retrieval performance. The fuzzy rules and fuzzy clusters are also useful for modeling user profiles that describe the groups of textual documents in which the user is interested. We apply fuzzy rules to adapt user queries to fuzzy inference within a sound and complete fuzzy logic system. We show some empirical results indicating that using our unified framework, the induction and application of fuzzy rules produces a more effective textual information retrieval system.

Keywords: Textual Information Retrieval, Fuzzy Rules, Rule Induction, Fuzzy Clustering, Query Expansion, User Profiles

In Szczepaniak, S., Segovia, J., Kacprzyk, J., and Zadeh, L.A. (eds.), Intelligent Exploration of the Web, Heidelberg, Germany: Physica-Verlag, 2002

Leemon Baird
Department of Computer Science, U.S. AIr Force Academy, USAFA, CO 80840
leemon.baird@usafa.edu

D.H. Kraft
Department of Computer Science
Louisiana State University
Baton Rouge, LA 70803-4020, USA

Abstract
The main objective of an information retrieval system is to be effective in providing a user with relevant information in response to a query. However, especially given the information explosion which has created an enormous volume of information, efficiency issues cannot be ignored. Thus, to be able to quickly process lists of documents that have the keywords stated in a given query assigned/indexed to them by merging via the Boolean logic of the query is essential in a Boolean query system. A new algorithm, based loosely on concurrent codes, is developed and discussed.