发明名称 |
DOCUMENT PROCESSING DEVICE AND PROGRAM |
摘要 |
PROBLEM TO BE SOLVED: To determine an appropriate cutout range of character strings according to classification in an expression composed of a plurality of words even if the classification to which an input sentence belongs is not known. SOLUTION: A morphological analysis part 32 extracts the plurality of words constituting a first character string included in the input sentence. An expression cutout 33 acquires a plurality of second character strings based on the plurality of words extracted by the morphological analysis part 32. A character string maintaining part 24 maintains the plurality of second character strings. A classification evaluation value calculation part 35 retrieves indices of the second character strings maintained in the character string maintaining part 24 from a document DB 22 by the classification. The classification evaluation value calculation part 35 calculates the evaluation value of each of the second character strings maintained in the character string maintaining part 24 for each classification based on the retrieved indices. A character string determination part 36 determines the second character strings of which evaluation values calculated by the classification evaluation value calculation part 35 satisfy a condition as the character strings cut out from the input sentence. COPYRIGHT: (C)2011,JPO&INPIT
|
申请公布号 |
JP2011039985(A) |
申请公布日期 |
2011.02.24 |
申请号 |
JP20090189280 |
申请日期 |
2009.08.18 |
申请人 |
TOSHIBA CORP;TOSHIBA SOLUTIONS CORP |
发明人 |
SAITO YOSHIMI;KANO TOSHIYUKI;NITTA SAORI;KOSHIBA ATSUSHI |
分类号 |
G06F17/30;G06F12/00 |
主分类号 |
G06F17/30 |
代理机构 |
|
代理人 |
|
主权项 |
|
地址 |
|