摘要
词语相似度计算是自然语言处理领域中的关键问题之一,在机器翻译、信息检索等方面有着重要的应用价值。在英文辅助写作系统中,因为缺少相关提示,用户起初往往不能明确自己的查询需求,导致不能快速而准确地检索到需要的信息,从而影响用户使用满意度。结合了语义词典WordNet和利用上下文信息对词语语义的约束性来区分语境变换带来的词语间相似度的差异的方法,提出了一种英文辅助写作系统中的相关提示词的生成方法,该方法生成优质的相关提示词,帮助用户快速且准确地检索到所需信息。
Word similarity is one key problem in Natural Language Processing, and it has important application value in machine translation, infor- mation retrieval, etc. In English auxiliary writing system, the users originally cannot define their inquiries due to absence of relevant tips. It causes the users not to quickly and accurately retrieve the needed information, thus effects customer satisfaction. Combining the semantic dictionary WordNet with the method that distinguish the differences of words similarity from context transform based on binding force of word semantic about the context information, the paper proposes a method of generating the related prompt word in English auxiliary writing system, and the method can generate high-quality related prompt word, which help the users quickly and accurately retrieve the needed information.
作者
占飞
刘挺
ZHAN Fei, LIU Ting (Computer Science & Technology School, Harbin Institute of Technology, Harbin 150001, China)
出处
《智能计算机与应用》
2011年第1X期51-54,58,共5页
Intelligent Computer and Applications
关键词
词语相似度
同义词推荐
相关提示词
语境
Word Similarity
Synonym Recommend
Related Prompt Word
Context