Abstract:
A method for automatic text segmentation and annotation is presented. It first discovers the themes presented in the document collection and then split each document according to these themes.
Keywords:text segmentation, natural language processing, information retrieval.
Fulltext:
PDF file (290 kB)