Abstract: According to the disadvantages of the method based on character string matching, the method based on comprehension and the method based on statistic, this paper presents a word segmentation method based on ontology and syntax analysis, which avoids the disadvantages about the lost of the semantic meaning with traditional method. It analyzes the grammar by building ontology, and inquires based on the intelligence. Experimental result proves that the accuracy is improved.

Key words: word segmentation, ontology, types of literature, syntax analysis

摘要： 针对基于字符串匹配的分词方法、基于理解的分词方法和基于统计的分词方法所存在的缺陷，提出基于本体和句法分析的某领域分词方法，通过建立体裁本体进行句法分析，从智能化的角度进行查词，避免了传统方法不考虑上下文信息导致的语义丢失等情况。实验结果证明，该方法可以较大地提高分词的精度。

关键词: 分词, 本体, 体裁, 句法分析

CLC Number: 

• TP391