【摘要】 词义消歧问题是自然语言处理方面的重点和难点之一。本文利用机器可读词典《知网》和《辞海》获取初始搭配资源,采用统计的方法对搭配集进行扩展;利用语言学中的搭配知识对扩展后的搭配集进行了有效的缩减;最后利用所获得的搭配集来对多义词进行词义消歧。经试验证明,该方法对词义消歧是有效的。更多还原

【Abstract】 WSD(Word Sense Disambiguation) is one of the difficult problems and a key point in natural language processing.The paper proposes the following WSD method.Machine-readable dictionaries,Hownet and Cihai,are imposed as the collocations resources,meanwhile a statistical method is used in the expanding.Then the expanding results are selected based on linguistics knowledge.Lastly,WSD is performed using the collocations got.The experiment result proves this method to be effective.更多还原