| 摘要 |
<p><P>PROBLEM TO BE SOLVED: To achieve text segmentation using Web retrieval for achieving text segmentation without requiring learning data. <P>SOLUTION: An input text is divided by sentence units, and the morphemic analysis of the divided sentences is carried out, and a noun, an adverb, a verb, an adjective, and an adjective verb whose morphemic analysis has been carried out are extracted as retrieval words, and the morphemic analysis of the text subjected to Web retrieval based on the retrieval words is carried out, and the noun, adverb, verb, adjective, and adjective verb among the analyzed morphemes are acquired as related words, and the connectivity of the plurality of sentences obtained by dividing the input text is determined by using a keyword group being the combination of the retrieval words and the related words, and the input text is divided by extracting logical paragraphs being sentences between valleys of the connectivity. <P>COPYRIGHT: (C)2009,JPO&INPIT</p> |
