Presented is an empirical confirmation of clustering behavior that has been proposed as a conjecture in the literature; namely, that document ordering is more significant in an algorithm using a full document space search, rather than a search of just unclustered documents, each time a density test is performed. (4 references) (Author)

This paper describes a set of four BASIC computer programs for computing the Rand index of cluster similarity. The routines calculate the unadjusted Rand Index; the separate routines are designed to handle problems of different sizes and alternate methods of data storage. (Author)

Johnson has shown that the single linkage and complete linkage hierarchical clustering algorithms induce a metric on the data known as the ultrametric. Johnson's proof is extended to four other common clustering algorithms. Two additional methods also produce hierarchical structures which can violate the ultrametric inequality. (Author/CTM)

A clustering technique is described, the objective of which is to detect deviant subpopulations which deviate from a primary subpopulation in a well defined direction. (Author/DEP)

A study was undertaken to classify mechanically a document collection using the free-language words in titles and abstracts of physics research papers. Using a clustering algorithm, results were obtained which closely duplicated clusters obtained by previous experiments with citations. A brief comparison is made with a traditional manual…

The classification capabilities of the one-dimensional Kohonen neural network (T. Kohonen, 1995) were compared with those of two partitioning and three hierarchical cluster methods in 2,580 data sets with known cluster structure. Overall, the performance of the Kohonen networks was similar to, or better than, that of the others. Implications for…

A clustering algorithm which is tree-like in structure, and is based on user queries, is presented. It is compared to some existing algorithms and is found to be superior. (Author)

Using examples of data from the areas of information retrieval and of multivariate data analysis, six hierarchic clustering algorithms (single link, median, centroid, group average, complete link, Wards's) are examined and evaluated by using three proposed coefficients of hierarchic structure. Nine references are cited. (EJS)

Using constraints with cluster analysis limits the possible number of clusters. This paper deals with clustering problems where grouping is constrained by a symmetric and reflexive relation. Two approaches, along with illustrations, are presented. (Author/JKS)

A general class of nonhierarchical clustering models and associated algorithms for fitting them are presented. These models generalize the Shepard-Arabie Additive clusters model. Two applications are given and extensions to three-way models, nonmetric analyses, and other model specifications are provided. (Author/JKS)

Monte Carlo validation studies of clustering algorithms, including Ward's minimum variance hierarchical method, are reviewed. Caution concerning the uncritical selection of Ward's method for recovering cluster structure is advised. Alternative explanations for differential recovery performance are explored and recommendations are made for future…

A method for hierarchical agglomerative polythetic (multivariate) clustering, based on unweighted pair group using arithmetic averages (UPGMA) is compared with the original beta-flexible technique, a weighted average method. Reasons the flexible UPGMA strategy is recommended are discussed, focusing on the ability to recover cluster structure over…

Discussion of models for information retrieval focuses on an application of linear algebra to text clustering, namely, a metric for measuring cluster quality based on the theory that cluster quality is proportional to the number of terms that are disjoint across the clusters. Explains term-document matrices and clustering algorithms. (Author/LRW)

One set of approaches to the problem of clustering with dichotomous data in cluster analysis (CA) was studied. The techniques developed for clustering with binary data involve calculating distances between observations based on the variables and then applying one of the standard CA algorithms to these distances. One of the groups of distances that…

A nonspatial operationalization of the Krumhansl distance-density model of similarity is presented. The conceptual model and empirical evidence are reviewed. A nonspatial, tree-fitting methodology is described, which is sufficiently flexible to fit several competing hypotheses of similarity formation. Extensions to spatial models, three-way…