Abstract: Several methods are proposed to automatically identify coordination relation, which is the most widely distributed one among discourse relations. The authors exploit semantic similarity and structure similarity to compute the sentence similarity, using lexical similarity, maximum common substring calculation, maximum length matching around head word, special words strengthening. Three of the above methods are integrated, and the experiment achieves promising results.

Key words: coordinate complex sentence, discourse relation, sentence similarity, ensemble method

摘要： 针对汉语句际关系中分布最广泛的并列复句, 提出一种自动识别的方法。通过对句子语义相似度和结构相似度的计算, 使用基于词义的句子相似度计算、最大公共子串、最大谓词周边匹配长度、加重特定词语复现等方法, 在广义并列关系上进行评测。最后将其中3种方法进行集成, 并取得了较为理想的效果。

关键词: 并列复句, 句际关系, 句子相似度, 集成方法