Matching method and system for related sentences in different documents and computer readable storage medium

The invention discloses a matching method for related sentences in different documents, which is used for matching reference sentences in reference documents with candidate sentences in comparison documents, and comprises the following steps: on three levels of shallow semantics, statistical information and deep semantics, calculating shallow-layer scores, statistical scores and deep-layer scoresof the candidate sentences relative to the reference sentences, wherein the shallow-layer scores, the statistical scores and the deep-layer scores represent matching degrees; and fitting the shallow score, the statistical score and the deep score based on a linear regression model to obtain a final score used for representing the matching degree of the candidate sentence relative to the referencesentence. The document matching accuracy is improved. 本发明公开了一种不同文档中相关句子的匹配方法，用于将基准文档中的基准句子和比对文档中的候选句子进行匹配，所述匹配方法包括：在浅层语义、统计信息、深层语义三个层面上，计算候选句子相对于基准句子的表示匹配程度的浅层分数、统计分数、深层分数；基于线性回归模型拟合所述浅层分数、统计分数和深层分数，获得用来表示所述候选句子相对于基准句子的匹配度的最终分