设计了一种定义抽取系统，通过混合使用模式规则、语法规则和词频统计以达到定义抽取的目的。详细介绍了系统中的各项关键技术，包括分词技术、抽取规则导入技术、句子抽取技术和加权词处理技术。并选取《情报学报》2007-2008年的235篇论文全文作为实验数据，通过与人工抽取结果进行对比分析，测试了抽取系统_的抽全率与抽准率。%This paper designs a definition extraction system, which reaches the purpose of definition extraction by the mixed use of pattern rules, grammar rules and word frequency statistics. The paper gives a detailed description of the key technologies of the system, including word segmentation, extraction rules import, sentence extraction and weighted word processing. Taking the 235 articles published in the ＂Journal of The China Society for Scientific and Technical Information＂ from 2007 to 2008 as experimental data, the paper tests the recall and precision of the extraction system by comparison with the artificially extracted results.
展开▼
