基于LSI的代码-文档可追溯关联挖掘研究

软件过程产品间可追溯关联挖掘对软件维护及需求跟踪等众多领域至关重要。基于此，提出一种基于潜在语义索引提取程序代码和中文文档关联信息的方法，该方法是对向量空间模型的改进，通过分析文本间隐含的语义结构来确定关联度，而不依赖于词项的匹配。实验结果表明，该方法不依赖于代码和文档预先定义的同义词库和知识库，并能一定程度上提高查全率和查准率。

Research on Code and Documentation Traceability Association Mining Based on LSI

Traceability link recovery among software process products is very important in many fields, such as software maintenance, as well as requirement trac. Based on Latent Semantic lndexing（LSI）, the traceability recovery information can be extracted automatically from program source code and the related Chinese documentation. The obvious advantage is that the presented method does not rely on the pre-defined thesaurus and knowledge for the code and documentation, and to some extent, it improves the recall and precision.