Self-Organising Maps for Tree View Based Hierarchical Document Clustering
By Richard Freeman, Hujun Yin and Nigel Allinson
Abstract
In this paper, we investigate the use of self-organising maps (SOMs) for document clustering. Previous methods using SOMs to cluster documents have used 2D maps. This paper presents a hierarchical and growing method using a series of 1D maps instead. Using this type of SOM is an efficient method for clustering documents and browsing them in a dynamically generated tree of topics. These topics are automatically discovered for each cluster, based on the set of documents in a particular cluster. We demonstrate the efficiency of the method using different sets of real-world Web documents
Keywords
self-organising maps, self-organising feature maps, SOM, contextual nformation, two-dimensional maps, topic hierarchies, content similarity
Bibliographic Details
@inproceedings{freemanIjcnno2,
Author = {Freeman, R. and Yin, Hujun and Allinson, N.M.},
Title = {Self-organising maps for tree view based hierarchical document clustering},
BookTitle = {Proceedings of 2002 International Joint Conference on Neural Networks (IJCNN),
12-17 May 2002},
Address= {Honolulu, HI, USA},
Publisher = {IEEE},
Volume = {Vol.2},
Pages = {1906-11},
Year = {2002} }
}