| 摘要 |
PROBLEM TO BE SOLVED: To highly precisely calculate a conceptual vector including a proper noun in language which does not have any conceptual base. SOLUTION: When the word of language B is not registered in a two language dictionary storage means, a meaning category name associated with the word of the language B is acquired by referring to a language B proper noun meaning category table storage means on the basis of the word of the language B, and the conceptual vector of the word of the language B is created by referring to a language A conceptual base storage means on the basis of the meaning category name. Furthermore, the created conceptual vector of the word of the language B is stored in a language B word conceptual base storage means, and when the document of the language B is input, the document is divided into words, and the appearance frequency of the divided words in the document is calculated, and the words are converted into vectors by referring to the language B word conceptual base storage means on the basis of the words, and a weighted average is calculated on the basis of the appearance frequency, and the weighted average is output as a document vector. COPYRIGHT: (C)2010,JPO&INPIT
|
