Title :
Cascade Chinese Potential Name Recognition
Author :
Hongjian, Liu ; Defeng, Guo ; Quan, Zhou ; Kenji, Nagamatsu ; Qinghua, Sun
Author_Institution :
Hitachi (China) R&D Corp., Shanghai, China
Abstract :
In this paper, cascade Chinese potential name recognition is proposed. Internal information of a person like family name, first name do not needed for name recognition while context keywords is used for name guessing. Some conceptions such as bidirectional potential name recognition, rough confirmation of potential name, segmentation word, cascade recognition and final person name confirmation is proposed. Experiment proves that the method presented in this paper have a special advantage in recognition on foreign translation name, special name, combinational name, ambiguity name, irregular name etc. The method can be taken as a pre-processing method of other person name recognition method with using internal name information. The method in the paper is not limit to Chinese name recognition but also can be generalized to other named entity identification.
Keywords :
computer vision; image recognition; natural languages; bidirectional potential name recognition; cascade Chinese potential name recognition; foreign translation name; named entity identification; preprocessing method; Information analysis; Information retrieval; Information technology; Laboratories; Speech recognition; Speech synthesis; Statistical analysis; Statistics; Sun; Viterbi algorithm; context keywords; named entity identification; potential name; segmentation word;
Conference_Titel :
Information Technology and Applications, 2009. IFITA '09. International Forum on
Conference_Location :
Chengdu
Print_ISBN :
978-0-7695-3600-2
DOI :
10.1109/IFITA.2009.85