| 摘要 |
PURPOSE:To exclude comparison on a character unit basis as much as possible and accurately detect an error position without requiring a complicated processing of clause units or the like at the time of comparison between an input document and a correct document. CONSTITUTION:Document data as an input model is stored in a comparison source document memory 4. A control part 1 compares document data inputted from an input part 2 with corresponding document data in the comparison source document memory 4. In this case, the control part 1 successively extracts character strings of continuity of the same character type from the head of both document data and discriminates whether extracted character strings coincide with each other or not to detect the error position of the document. |
