A Method about Construction of Chinese Digital Dictionary
For the increasing of electronic text and the creating of new words on the Internet continuously, how to construct and renew a lexicon is very important research. Through investigating those present methods of Chinese word segmentation, proposed a hybrid method for constructing a lexicon of open text. The method makes use of the merit of rules, Inductive Learning and probabilistic model, combines various techniques together, and improved the Inductive Learning Method. The whole processing is a half-supervise process that needs a little manual labor. It would be an assistance means that construct an electronic lexicon, would be lower the manual labor consumedly and improve efficiency of lexicon construction. The usefulness of the different strategy in the method is proved by the different experiment. The experiment result shows the validity of the method.
rules open text semi-supervised learning digital dictionary construction
Zhongjian Wang Ling Wang
School of Computer and Information Engineering, Harbin University of Commerce, China School of Foreign Language, Harbin University of Commerce, China
国际会议
2010 International Conference on Circuit and Signal Processing(2010年电路与信号处理国际会议 ICCSP 2010)
上海
英文
330-333
2010-12-25(万方平台首次上网日期,不代表论文的发表时间)