摘要 |
PROBLEM TO BE SOLVED: To enable character mapping in a character string set of different languages to be performed with high accuracy.SOLUTION: Concerning a character string set which is a combination of character strings having the same meaning, each belonging to different first and second languages, when each character string in the character string set is to be configured, in order from the beginning of the character strings, with a pre-transliteration segment, a transliteration segment, and a post-transliteration segment, a mapping calculation unit 23 configures each character string in the character string set with the pre-transliteration segment, the transliteration segment, and the post-transliteration segment on the basis of a transliteration model selection probability, non-transliteration model selection probability, non-transliteration model generation probability, and transliteration model generation probability, and maps characters between a partial character string of the transliteration segment among the character strings of the first language and a partial character string of the transliteration segment among the character strings of the second language to each other so that the character string set becomes plausible. |