Two partitioning type clustering algorithms

Can, Fazlı; Ozkarahan, Esen A.

doi:10.1002/asi.4630350503

Cited by 19 publications

(21 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This inequality, stated for the appearance of a term in a centrold, will emphasize the terms which appear often enough within a cluster and having noticeable uniqueness. (It is shown [3,5] that the above condition will provide the appearance of the terms having a uniqueness value greater than or equal to the average decoupling coefficient of the terms in at least one eentroid.) ]n Algorithm-2, a term is assigned to a eentroid if the term appears in at least T (a threshold value) members of the corresponding cluster.…”

Section: The Clustering Algorithmmentioning

confidence: 98%

“…Also, rain(t) = max(re,n) holds if each term is assigned to only one document and the document ls described by more then one term (i.e., n>m) or each term is assigned to more than one document and all documents are described by only one term (i.e., m > n). Therefore, it was previously shown [3,5] that the number of clusters for the documents and the terms used for the description of documents are equal to each other. Furthermore, since our clustering is of partitioning type, n= cannot be greater than rain(re,n) (which in the foregoing was shown to be equal to either m or n), for otherwise the partitioning property would be violated.…”

Section: Ingmentioning

confidence: 99%

“…The average complexity of the algorithms is 0(rn, Z/log m) and 0(k x rn,2/ log m) for Algorithm-I and Algorithm-2, respectively, where k is the number of iterations [2,3,5].…”

Section: The Clustering Algorithmmentioning

confidence: 99%

“…In this process, some seeds might be identical, since they may be described by nearly identical terms. An algorithm to identify and eliminate the identical (false) seeds has been introduced [2, 3,5].…”

Section: The Clustering Algorithmmentioning

confidence: 99%

“…In the remainder of this article, we will discuss some newly introduced concepts of the partitioning type document and term clustering methodology we have introduced earlier [2,3.4,5]. A detailed description of the so-called covet coefficient based clustering algorithm appeared in [3]. The experiments conducted to analyze the stability and similarity behaviour of the algorithms that are based on the cover coefficient concept have recently appeared in [4].…”

Section: Introductionmentioning

confidence: 98%

See 4 more Smart Citations

Concepts of the cover coefficient-based clustering methodology

Can

Ozkarahan

1985

Proceedings of the 8th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval - SIGIR

Self Cite

View full text Add to dashboard Cite

Document cLzaste~ng has several unresolved. problems. Among them are high time and space complexity, difficulty of determining similarity thresholds, order dependence, nonuniform document distribution in clusters, and arbitrariness in determination of various cluster initiators. To overcome these problems, to some degree, the cover coefficient based clustering methodology has been introduced. The concepts used in this methodology have created certa/n nezv concepts, relationships, and meavures such as the effect of indexing on clustering, an optimal vocabulary generation for indezing, and a ne~v matching function.These new concepts are discussed. The result of performance experiments that shoe the effectiveness of the clustering methodology. and the matching function are also included. In these experiments, ~t has been also observed that the majority of the documents obtained in a search are concentrated in a fee clusters containing a lozu percentage of documents of the database.

show abstract

Section: The Clustering Algorithmmentioning

confidence: 98%

Section: Ingmentioning

confidence: 99%

“…The average complexity of the algorithms is 0(rn, Z/log m) and 0(k x rn,2/ log m) for Algorithm-I and Algorithm-2, respectively, where k is the number of iterations [2,3,5].…”

Section: The Clustering Algorithmmentioning

confidence: 99%

Section: The Clustering Algorithmmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 98%

See 3 more Smart Citations

Concepts of the cover coefficient-based clustering methodology

Can

Ozkarahan

1985

Proceedings of the 8th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval - SIGIR

Self Cite

View full text Add to dashboard Cite

show abstract

Computation of term/document discrimination values by use of the cover coefficient concept

Can

Ozkarahan

1987

J. Am. Soc. Inf. Sci.

Self Cite

View full text Add to dashboard Cite

Indexing in information retrieval (IR) is used to obtain a suitable vocabulary of index terms and optimum assignment of these terms to documents for increasing the effectiveness and efficiency of an IR system. The concept of term discrimination value (TDV) is one of the criteria used for index-term selection. In this article a new concept called the cover coefficient (CC) will be used in computing TDVs. After a brief introduction to the theory of indexing and the CC concept, an efficient way of computing TDVs by use of the CC concept, index-term selection , and weight modification are discussed. It is also shown that the computational cost of the CC approach in the calculation of TDVs is favorably comparable to the cost of a different approach that uses similarity coefficients. Furthermore, the TDVs obtained by the CC approach are consistent with those of the latter approach.

show abstract

High‐speed rough clustering for very large document collections

Kishida

2010

J. Am. Soc. Inf. Sci.

View full text Add to dashboard Cite

Document clustering is an important tool, but itis not yet widely used in practice probably because of its high computational complexity. This article explores techniques of high-speed rough clustering of documents, assuming that it is sometimes necessary to obtain a clustering result in a shorter time, although the result is just an approximate outline of document clusters. A promising approach for such clustering is to reduce the number of documents to be checked for generating cluster vectors in the leader-follower clustering algorithm. Based on this idea, the present article proposes a modified Crouch algorithm and incomplete single-pass leaderfollower algorithm. Also, a two-stage grouping technique, in which the first stage attempts to decrease the number of documents to be processed in the second stage by applying a quick merging technique, is developed. An experiment using a part of the Reuters corpus RCV1 showed empirically that both the modified Crouch and the incomplete single-pass leader-follower algorithms achieve clustering results more efficiently than the original methods, and also improved the effectiveness of clustering results. On the other hand, the two-stage grouping technique did not reduce the processing time in this experiment.

show abstract

Two partitioning type clustering algorithms

Cited by 19 publications

References 4 publications

Concepts of the cover coefficient-based clustering methodology

Concepts of the cover coefficient-based clustering methodology

Computation of term/document discrimination values by use of the cover coefficient concept

High‐speed rough clustering for very large document collections

Contact Info

Product

Resources

About