摘要 |
PROBLEM TO BE SOLVED: To realize an accurate and highly responsive retrieval means in original-document retrieval in which paper documents are scanned and original electronic documents are retrieved. SOLUTION: Texts read by an OCR are classified into similar character groups in which characters are preliminarily grouped according to morphological similarities, and character codes are converted to characters representative of the similar character groups to search similar documents. By this, an accurate original-document retrieval is realized, which is not affected by erroneous identification of a fine symbol such as minus or dash. COPYRIGHT: (C)2007,JPO&INPIT
|