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)

Substantial alterations to a system often have little or no effect on particular collections. This may be due to poor separation of relevant and non-relevant documents. The paper presents a procedure for characterizing this separation, which can show whether proposed modifications of the base system are likely to be useful. (8 references)…

Retrieval performance with automatic term classifications for three test collections has been variable. This paper attempts to discover why. The real difference between the collections is in the separation of relevant from non-relevant documents. The separation is so poor that classification cannot be expected to succeed. (14 references)…

Statistical clustering is critical in designing scalable image retrieval systems. This paper presents a scalable algorithm for indexing and retrieving images based on region segmentation. The method uses statistical clustering on region features and IRM (Integrated Region Matching), a measure developed to evaluate overall similarity between images…

Proposes a composite feature measure which combines the shape and color features of an image based on a clustering technique. A similarity measure computes the degree of match between a given pair of images; this technique can be used for content-based image retrieval of images using shape and/or color. Tests the technique on two image databases;…

Defines relationship between user profile and user query in terms of relationship between clusters of documents retrieved by each, and explores the expression of cluster similarity and cluster overlap as linear functions of similarity existing between original pairs of profiles and queries, given the desired retrieval threshold. (23 references)…

Describes a study conducted on the cystic fibrosis (CF) database, a subset of MEDLINE, that investigated clustering structure and the effectiveness of cluster-based retrieval as a function of the exhaustivity of the uncontrolled subject descriptions. Results are compared to calculations for controlled descriptions based on Medical Subject Headings…

Discusses the application of theories of cognitive modeling to information systems design and describes research that investigated the validity of the repertory grid for incorporation into intelligent front-end interfaces for information storage and retrieval systems. Personal construct theory is discussed and future research is suggested. (67…

This study attempts to analyze descriptors taken from subject categories in ERIC thesaurus and used as search terms on CROSS database Bibliographic Retrieval Services. An expectation ratio was computed and cluster analysis was conducted to discover subject relationships among databases. A list of databases retrieved and 12 references are appended.…

Examines the use of various statistical techniques for generating domain-oriented multiword vocabulary terms for natural language database systems. Conclusions show the vocabulary clustering effect should be considered when making significance calculations and that a simple ratio of subject matter relative frequency to total sample relative…