Abstract. Fast and high-quality document clustering algorithms play an important role in providing intuitive
navigation and browsing mechanisms by organizing large amounts of information into a small number of meaningful
clusters. In particular, clustering algorithms that build meaningful hierarchies out of large document collections
are ideal tools for their interactive visualization and exploration as they provide data-views that are consistent,
predictable, and at different levels of granularity. This paper focuses on document clustering algorithms that build
such hierarchical solutions and (i) presents a comprehensive study of partitional and agglomerative algorithms
that use different criterion functions and merging schemes, and (ii) presents a new class of clustering algorithms
called constrained agglomerative algorithms, which combine features from both partitional and agglomerative
approaches that allows them to reduce the early-stage errors made by agglomerative methods and hence improve
the quality of clustering solutions. The experimental evaluation shows that, contrary to the common belief,
partitional algorithms always lead to better solutions than agglomerative algorithms; making them ideal for clustering
large document collections due to not only their relatively low computational requirements, but also higher
clustering quality. Furthermore, the constrained agglomerative methods consistently lead to better solutions than
agglomerative methods alone and for many cases they outperform partitional methods, as well.
%0 Journal Article
%1 zhao05hierarchical
%A Zhao, Ying
%A Karypis, George
%A Fayyad, Usama
%D 2005
%I Kluwer Academic Publishers
%J Data Mining and Knowledge Discovery
%K clustering hierarchical
%N 2
%P 141--168
%R 10.1007/s10618-005-0361-3
%T Hierarchical Clustering Algorithms for Document Datasets
%U http://dx.doi.org/10.1007/s10618-005-0361-3
%V 10
%X Abstract. Fast and high-quality document clustering algorithms play an important role in providing intuitive
navigation and browsing mechanisms by organizing large amounts of information into a small number of meaningful
clusters. In particular, clustering algorithms that build meaningful hierarchies out of large document collections
are ideal tools for their interactive visualization and exploration as they provide data-views that are consistent,
predictable, and at different levels of granularity. This paper focuses on document clustering algorithms that build
such hierarchical solutions and (i) presents a comprehensive study of partitional and agglomerative algorithms
that use different criterion functions and merging schemes, and (ii) presents a new class of clustering algorithms
called constrained agglomerative algorithms, which combine features from both partitional and agglomerative
approaches that allows them to reduce the early-stage errors made by agglomerative methods and hence improve
the quality of clustering solutions. The experimental evaluation shows that, contrary to the common belief,
partitional algorithms always lead to better solutions than agglomerative algorithms; making them ideal for clustering
large document collections due to not only their relatively low computational requirements, but also higher
clustering quality. Furthermore, the constrained agglomerative methods consistently lead to better solutions than
agglomerative methods alone and for many cases they outperform partitional methods, as well.
@article{zhao05hierarchical,
abstract = {Abstract. Fast and high-quality document clustering algorithms play an important role in providing intuitive
navigation and browsing mechanisms by organizing large amounts of information into a small number of meaningful
clusters. In particular, clustering algorithms that build meaningful hierarchies out of large document collections
are ideal tools for their interactive visualization and exploration as they provide data-views that are consistent,
predictable, and at different levels of granularity. This paper focuses on document clustering algorithms that build
such hierarchical solutions and (i) presents a comprehensive study of partitional and agglomerative algorithms
that use different criterion functions and merging schemes, and (ii) presents a new class of clustering algorithms
called constrained agglomerative algorithms, which combine features from both partitional and agglomerative
approaches that allows them to reduce the early-stage errors made by agglomerative methods and hence improve
the quality of clustering solutions. The experimental evaluation shows that, contrary to the common belief,
partitional algorithms always lead to better solutions than agglomerative algorithms; making them ideal for clustering
large document collections due to not only their relatively low computational requirements, but also higher
clustering quality. Furthermore, the constrained agglomerative methods consistently lead to better solutions than
agglomerative methods alone and for many cases they outperform partitional methods, as well.},
added-at = {2007-02-12T01:06:37.000+0100},
author = {Zhao, Ying and Karypis, George and Fayyad, Usama},
biburl = {https://www.bibsonomy.org/bibtex/29e00e2aa62763eacf609cb66171c965b/marianne},
citeulike-article-id = {196574},
doi = {10.1007/s10618-005-0361-3},
interhash = {fdb19684fd849b0c44ac0fb996b7888e},
intrahash = {9e00e2aa62763eacf609cb66171c965b},
issn = {1384-5810},
journal = {Data Mining and Knowledge Discovery},
keywords = {clustering hierarchical},
month = {March},
number = 2,
pages = {141--168},
priority = {2},
publisher = {Kluwer Academic Publishers},
timestamp = {2007-02-12T01:06:37.000+0100},
title = {Hierarchical Clustering Algorithms for Document Datasets},
url = {http://dx.doi.org/10.1007/s10618-005-0361-3},
volume = 10,
year = 2005
}