摘要
提出了利用知网的相关知识,结合分词、词语相关度计算的方案解决可拓策略生成系统对矛盾问题的界定与建模时的语义二义性的问题,尝试提高计算机语言表达用户信息的准确度和可拓策略生成系统解决矛盾问题的成功率.通过实验表明,知网与可拓策略生成系统的综合应用,可提高可拓策略生成系统解决矛盾问题的准确性和效率.
It proposes the use of related knowledge of HowNet,combined with the word segmentation,word correlation computing to solve the problems with extension definition of contradiction and ambiguity of semantic modeling of the strategy generating system,tries to improve computer language to express the user information accuracy and the extension strategy generating system and to increase the success rate.Experimental results show that the comprehensive application of HowNet and the extension strategy generating system makes it possible to improve the accuracy and efficiency of the extension strategy generating system in resolving contradictions.
出处
《广东工业大学学报》
CAS
2014年第2期21-26,共6页
Journal of Guangdong University of Technology
基金
国家自然科学基金资助项目(61273306)
关键词
矛盾问题
可拓学
知网
相关度
contradiction
Extenics
HowNet
relevance degree
