Presents a genetic relevance optimization process performed in an information retrieval system that uses genetic techniques for solving multimodal problems (niching) and query reformulation techniques. Explains that the niching technique allows the process to reach different relevance regions of the document space, and that query reformulations…

Discussion of semantic networks and ranked retrieval focuses on two models, the semantic network model with spreading activation and the vector space model with dot product. Suggests a formal method to analyze the two models in terms of their relative performance in the same universe of objects. (Author/LRW)

Presents an approach to a Genetic Information Retrieval Agent Filter (GIRAF) that filters and ranks documents retrieved from the Internet according to users' preferences by using a Genetic Algorithm and fuzzy set theory to handle the imprecision of users' preferences and users' evaluation of the retrieved documents. (Author/LRW)

Proposes local similarity pattern (LSP) as a new method for computing digital image similarity. Topics include optimizing similarity computation based on genetic algorithm; relevance feedback; and an evaluation of LSP on five databases that showed an increase in retrieval precision over other methods for computing image similarity. (Author/LRW)

Presents a model of information retrieval based on plausible inference which suggests that techniques should be found for combining multiple search strategies into an overall assessment of a document's relevance. The results of experiments designed to test this approach using a simple spreading activation search are discussed. (21 references) (CLB)

Proposes a novel approach to automatically retrieve keywords and then uses genetic algorithms to adapt the keyword weights. Discusses Chinese text retrieval, term frequency rating formulas, vector space models, bigrams, the PAT-tree structure for information retrieval, query vectors, and relevance feedback. (Author/LRW)

Discussion of query processing focuses on techniques developed by LEXIS-NEXIS for identifying key concepts, selecting synonyms and related terms, and detecting phrases that can be combined to increase recall and precision. Describes experiments in the ad hoc track of the sixth TREC (Text Retrieval Conference), including the use of ranking…

This study examined the effectiveness using an automatic algorithm to rank the results of Boolean searches of an inverted file design document retrieval system. Relevant documents were ranked significantly higher than nonrelevant documents on output lists. (Author/KP)

Describes a computerized information retrieval system that uses three agglomerative hierarchic clustering algorithms--single link, complete link, and group average link--and explains their implementations. It is noted that these implementations have been used to cluster a collection of 12,000 documents. (LRW)

Introduces two distribution schemes that partition documents over multiple processors and the corresponding multiprocessor retrieval algorithms that match relevant documents to user queries. The suggested framework is based on a general purpose hypercube multicomputer architecture with a dedicated disk for each node. A timing analysis for the…

Describes an approach that combines concepts from information retrieval, fuzzy set theory, and genetic programing to improve weighted Boolean query formulation via relevance feedback. Highlights include background on information retrieval systems; genetic algorithms; subproblem formulation; and preliminary results based on a testbed. (Contains 12…

Presents a relevance-feedback strategy that improves the effectiveness of information-retrieval systems based on back-propagation of the relevance of retrieved documents using an algorithm developed in a neural approach. Describes a neural information-retrieval model and reports results obtained with the algorithm in three different environments.…

Investigates methods for the correct and efficient handling of annotations in a full-text information-retrieval system in a hypertext environment. Goals were to increase relevant items retrieved, reduce non-relevant items, permit selective retrieval, and to explore the notion of distance between words as given in the distance-evaluation algorithm.…

Discussion of electronic news available on the World Wide Web focuses on a methodology of associating news documents using similarity measures and a name-phrase algorithm to create news representations that can be compared and ranked to find related news items. Considers precision over recall and interactivity with the users. (Author/LRW)

Presents a new stemming algorithm based on tree structures that improves relevance in information retrieval by conflation, grouping similar words into a single term. Highlights include the normalization process used in automatic thesaurus construction; theoretical aspects; the normalization algorithm; and experiments with English and Spanish. (LRW)