WORD VECTOR GENERATION DEVICE, WORD VECTOR GENERATION METHOD, PROGRAM, AND RECORDING MEDIUM WITH PROGRAM RECORDED THEREIN

摘要

PROBLEM TO BE SOLVED: To provide a word vector generation device for calculating relative weight showing an appearance frequency of a word in different genres.

SOLUTION: This word vector generation device includes: an appearance frequency calculation means for receiving a set of genres and a set of documents associated with each genre, and calculating the appearance frequency of each word in the set of documents associated with each genre; an appearance probability calculation means for normalizing the appearance frequency of each word obtained by the appearance frequency calculation means for each genre, and calculating the appearance probability of each word in each genre ; and a first word vector generation means for normalizing the appearance probability obtained by the appearance probability calculation means for each word to calculate its relevance with each genre, and for generating a first word vector whose coordinates are associated with the genre, and whose value shows its relevance with the genre.

COPYRIGHT: (C)2009,JPO&INPIT