The experiments involved 11,571 abstracts (with titles), 1,000 key-word stems and 93 search requests. Measures of word association are derived in several ways from the numbers of documents in which two given words co-occur, and measures of similarity from the numbers of words associated with both. Word clusters with different degrees of overlap…

Presents system for automatic handling of ordered sets, states based on these sets, and differing points of view regarding Universe of Discourse. Aspects are represented by new logical "overlap" function with examples taken from Ranganathan's horse and carriage parable and several books involving four main concepts (history, geography,…

Describes a cluster-based information retrieval procedure that can significantly reduce the computational requirements of the single linkage method, while still maintaining the retrieval effectiveness of the resulting classifications. Use of nearest neighbors, experimental details, and results and conclusions are highlighted. Fourteen references…

Describes a method of iterative searching that uses local metrical clustering in the formulation of additional searches by producing searchonyms, terms which are synonymous to keywords in the context of a single query. The method was applied to a database containing Hebrew and Aramaic text. Seven references are listed. (FM)

The single-link cluster method is used to construct a hierarchic classification for the 1400 documents in the Cranfield test collection. A variety of retrieval strategies applied to this hierarchy are evaluated in terms of effectiveness and efficiency. (Author)

Reports the results of experiments in document clustering using three well-known test collections. (Author)

Describes ways of using Boolean questions with a clustered information file, and reviews an experiment studying the effect of such a retrieval method. (FM)

Sorting things into groups is a basic intellectual task that allows people to simplify with minimal reduction in information. Classification techniques, which include both clustering and discrimination, provide step-by-step computer-based procedures for sorting things based on notions of generalized similarity and on the "class description"…

Two papers are included as Part Four of this report on Salton's Magical Automatic Retriever of Texts (SMART) project report. The first paper: "A Controlled Single Pass Classification Algorithm with Application to Multilevel Clustering" by D. B. Johnson and J. M. Laferente presents a single pass clustering method which compares favorably…

Describes the use of relevance feedback to select additional search terms and discusses the extraction of these terms from a maximum spanning tree connecting all terms in the index term vocabulary; retrieval effectiveness for different spanning trees is shown to be similar. Eight references are included. (Author/BK)

Proposes a means for determining the similarity between search request formulations in online information retrieval systems, and discusses the use of similarity measures for clustering search formulations and document files in such systems. Experimental results using the proposed methods are presented in three tables. A reference list is provided.…

Defines the notion of serial clustering of words in text, and explores the value of such clustering as an indicator of a word's bearing content. This approach is flexible in the sense that it is context-sensitive; a numerical approach may also be of value in assigning weights to terms in requests. Experimental support is obtained from natural text…

Discusses precision and recall in information science and proposes a new model based on fractal geometry for clusters of relevant documents. Search strategies for retrieving a group of relevant documents are reviewed; fractal sets and chaotic processes are described; and the new model is explained. (Contains 43 references.) (LRW)

Reports on comparative study of document classifications produced by use of single linkage, complete linkage, group average, and Ward clustering methods. Findings of work that compares use of clusters consisting of pairs of documents with conventional best match searches are also reported. Thirty-four references are provided. (EJS)

Considers classifications produced by application of single linkage, complete linkage, group average, and word clustering methods to Keen and Cranfield document test collections, and studies structure of hierarchies produced, extent to which methods distort input similarity matrices during classification generation, and retrieval effectiveness…