Sentence classification program, method, and sentence analysis server for classifying sentences containing unknown words

摘要

PPROBLEM TO BE SOLVED: To provide a text sort program which can sort even if text information described by colloquial "relaxed and informal expression" into a proper category, and to provide a method and a text analysis server. PSOLUTION: The text sort program includes: a reference text storage means which sorts a plurality of pieces of reference text information into the plurality of categories and stores them; an unknown word extraction means which divides object text information of sorting object containing unknown words into morphemes, and extracts unknown words; a search key generation means to generate a search key which includes the unknown word and the front and backward adjacent morphemes; a correction rule search means which searches one or more correction rule candidates which resembles the unknown word using the search key for each category of reference text information; a correction rule candidate selection means which selects one correction rule candidate based on text composition index for each category of the reference text information; a category selection means which selects the category of highest score based on text composition index of the correction rule; and a text sorting means to sort the object text information into the selected categories. PCOPYRIGHT: (C)2011,JPO&INPIT