An Integrated Approach to Information Retrieval with Fuzzy Clustering and Fuzzy Inferencing
Jianhua Chen, A. Mikulcic, Donald H. Kraft · Studies in fuzziness and soft computing · 2000
We present an integrated approach to fuzzy information retrieval which combines techniques in fuzzy set theory with methodologies in textual retrieval in order to achieve optimal retrieval performance. To capture the relationships among index terms, fuzzy logic rules (with truth value assignment in the 0–1 interval) are used. We adapt several fuzzy clustering methods (such as fuzzy c-means and fuzzy hierarchical clustering) to the task of clustering documents with respect to the terms. The clusters generated provide a basis for building the fuzzy logic rules. The clusters can also be used to form hyperlinks between documents. A previously developed fuzzy logic system, found to be sound and complete, is applied for fuzzy inferencing to derive useful modifications of the initial query, which will guide the search for relevant documents. Thus, this method combines fuzzy inference with traditional relevance feedback approach for retrieval. The advantage of this method is in the emphasis on semantic information (embodied in the rules and the inference mechanisms) which should lead to superior performance. A series of experiments conducted in order to validate this approach are presented, along with results and conclusions. These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.