关键词: 需求跟踪, 动态需求跟踪, 语义聚类, 句法分析, 自然语言处理

Abstract: Software requirement traceability has been globally recognized as a key factor of affecting the success of software projects. The generated traceability links are vital to many software engineering and software verification and validation (V&V) activities such as change impact analysis, software reuse and consistency checking. Addressing most existing requirement traceability approaches based on information retrieval are strongly affected by the quality of the documentation of different types of software artifacts, this paper presents a dynamic requirement traceability approach based on syntactic analysis. The proposed approach is able to extract terms which are most likely to characterize itself from text-based software artifacts such as source code and requirement document, and then reduce the adverse effects of noise in artifacts to the requirement tracing process. In order to verify the effectiveness of the proposed approach, we have compared the quality of the trace links produced by several dynamic requirement traceability approaches on three open source software systems and six types of software artifacts. The result suggests that the dynamic requirement traceability approach based on syntactic analysis can effectively improve the accuracy of the produced trace links.

Key words: trace requirement, dynamic requirement traceability, semantic cluster, syntactic analysis, natural language process