Adaptation of Hierarchical Clustering by Areas for Automatic Construction of Electronic Catalogue
Fedor Borisyuk, Vladimir Shvetsov · Proceedings of the Spring/Summer young researchers' colloquium on software engineering · 2010
This paper explores adaptation of Hierarchical clustering by areas algorithm, presented in [1], for automatic construction of electronic catalogue of text documents.Electronic catalogue is traditionally maintained and built by hand, significant research was done to classify the documents to existing hierarchical structures, but little was done about to build the electronic catalogue from scratch.We describe enhancements of Hierarchical clustering by areas algorithm and compare it (to be applied in the field of automatic catalogue construction) with other researches.This paper also proposes an algorithm of feature space reduction for preparation of document representations for text clustering.