Mining text documents for thematic hierarchies using self-organizing maps

摘要: Recently, many approaches have been devised for mining various kinds of knowledge from texts. One important application text is to identify themes and the semantic relations among these categorization. Traditionally, were arranged in a hierarchical manner achieve effective searching indexing as well easy comprehension human beings. The determination category their structures was mostly done by experts. In this work, we developed an approach automatically generate reveal structure them. We also used generated categorize documents. document collection trained self-organizing map form two feature maps. then analyzed maps obtained structure. Although test corpus contains documents written Chinese, proposed can be applied any language, such transformed into list separated terms.