Proposes a method to improve retrieval performance of the vector space model by using users' relevance feedback. Discusses the use of singular value decomposition and the latent semantic indexing model, and reports the results of two experiments that show the effectiveness of the proposed method. (Author/LRW)

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.…

Describes a study that computed low performance standards for the group of queries in 13 information retrieval (IR) test collections. Derived from the random graph hypothesis, these standards represent the highest levels of retrieval effectiveness that can be obtained from meaningless clustering structures. (Author/LRW)

Discussion of connectionist views for adaptive clustering in information retrieval focuses on a connectionist clustering technique and activation spreading-based information retrieval model using the interaction information retrieval method. Presents theoretical as well as simulation results as regards computational complexity and includes…

Describes a model of an information retrieval system that is based on a document re-ranking method, using document clusters. Retrieves documents based on the inverted file method, then analyzes the retrieved documents using document clusters and re-ranks them. Shows significant improvements over the method based on similarity search ranking alone.…

This two-part article combines a comprehensive account of a probabilistic model of retrieval with new systematic experiments on TREC (Text Retrieval Conferences) Program material. Part 1 covers the foundations and the model development for document collection and relevance data, along with the test apparatus. Data and results tables for both parts…

A comprehensive account of a probabilistic model of retrieval with new systematic experiments on TREC (Text Retrieval Conferences) Program material. Part 2 covers the further development of the model, with testing, and briefly considers other environment conditions and tasks, model training, concluding with comparisons with other approaches and an…

Describes a study that computed the low performance standards for queries in 17 test collections. Predicted by the hypergeometric distribution, the standards represent the highest level of retrieval effectiveness attributable to chance. Operational levels of performance for vector-space and other retrieval models were compared to the standards.…

Describes a theoretic process to apply Bahadur-Lazarsfeld expansion (BLE) to general probabilistic models and the state-of-the-art 2-Poisson model. Through experiments on two standard document collections, one in Korean and one in English, it is demonstrated that incorporation of term dependences using BLE significantly contributes to performance…

Discussion of relevance in information retrieval focuses on manifestations of relevance within a system of relevance. Discusses intention, context, inference, and interaction; explains the concepts of affective, motivational, situational, and socio-cognitive relevance; and proposes a consolidated model of relevance manifestations. (Author/LRW)

Explains how a competition-based connectionist model for diagnostic problem-solving is adapted to information retrieval. Topics include probabilistic causal networks; Bayesian networks; the neural network model; empirical studies of test collections that evaluated retrieval performance; precision results; and the use of a thesaurus to provide…

Discussion of user-relevance judgments in information retrieval focuses on findings from four separate studies of relevance judgments by 55 users conducting their initial online search on a particular information problem. Highlights include a theoretical framework, models of relevance, and implications for information-retrieval system design and…

Discussion of information retrieval and relevance focuses on mutual information, a measure which represents the relation between two words. A model of a natural-language information-retrieval system that is based on a two-level document-ranking method using mutual information is presented, and a Korean encyclopedia test collection is explained.…

Presents an interpolation theorem for an extended Boolean information retrieval model. Results show that whenever two or more documents are similarly ranked at any two points for a query containing exactly two terms, then they are similarly ranked at all points in between; and that results can fail for queries with more than two terms. (Author/LRW)

Discussion of the use of Latent Semantic Indexing to determine relevancy in information retrieval focuses on statistical regression and Bayesian methods. Topics include keyword searching; a multiple regression model; how the regression model can aid search methods; and limitations of this approach, including complexity, linearity, and…