CASE COMPONENT EXTRACTION PROGRAM

摘要

PROBLEM TO BE SOLVED: To extract a case component from an extracted character string with a high degree of accuracy.SOLUTION: A case component extraction program causes a computer to execute: a preprocessing step of extracting character strings from descriptions in a scope of claim for patent in a patent specification converted into electronic data and extracting at least a title of the invention from the character strings; a word packaging step of dividing the character strings into words by applying a morphological analysis and a syntax analysis to the character strings processed in the preprocessing step and regarding a prescribed character string as one word; a noun phrase extraction step of extracting noun phrases from the character string including the words after the word packaging step; and a case component extraction step of extracting a noun phrase that conforms to a prescribed rule among the noun phrases excepting the noun phrase identical to the title of the invention, as a case component, from the noun phrases extracted in the noun phrase extraction step.