DOCUMENT IDENTITY EXTRACTION DEVICE, DOCUMENT IDENTITY EXTRACTION METHOD, DOCUMENT CLASSIFICATION DEVICE, DOCUMENT CLASSIFICATION METHOD, DOCUMENT RETRIEVAL DEVICE, DOCUMENT RETRIEVAL METHOD, COMPUTER PROGRAM, AND RECORDING MEDIUM WHERE COMPUTER PROGRAM IS RECORDED

摘要

PROBLEM TO BE SOLVED: To provide a document identity extraction device and a document identity extraction method that have high precision in classifying or retrieving a document using machine learning and extract an identity of an object document, and a document classification device and a document retrieval device that use the identity of the document.;SOLUTION: A document identity extraction device generates a word string by connecting adjacent words in a specific region and uses presence of word strings or words together in the specific region, i.e. co-occurrence as an identity of an object document.;SELECTED DRAWING: Figure 8;COPYRIGHT: (C)2017,JPO&INPIT