CHARACTER STRING SPECIFYING SYSTEM

PURPOSE:To identify an omitted description and a formal description as the description indicating the same word concerning a document in which the formal description and omitted description of the same proper noun are mixed and repeatedly appear. CONSTITUTION:An input character string is divided into words for the unit of a sentence and a part of speech is applied to an input document by morpheme analystic processing 11. Next, the word with high possibility to be a proper noun (proper noun candidate) is selected out of the words in the document based on a part of speech applied by the morpheme analystic processing, and registered on the associated arrangement of a memory by proper noun candidate selection processing 12. Afterwards, two candidates are extracted from the proper noun candidates in the associated arrangement by omitted proper noun decision processing 13 and both the candidates are compared. When the common character string in the same order more than two characters exists between both of them and that common character string is provided with the same number of characters as one candidate at least, those two candidates are judged as the descriptions of the same proper noun and written in the associated arrangement after applying the same identification code to both of them. Concerning all the pairs of candidates to be combined, the omitted proper noun decision processing 13 repeats this processing.