An Automatic Approach to Extracting Requirement Dependencies based on Semantic Web

For a complex system with a large number of requirement texts, manual analysis of the dependencies between the requirement texts is very time-consuming and requires heavy workload. This paper proposes an automatic approach to extract requirement dependencies based on semantic web. Firstly, requirements text is formalized as a set of keywords by segmenting the requirement text and making part-of-speech tagging. Secondly, the requirement dependencies are automatically obtained by constructing the semantic web ontology and defining inference rules. Finally, the results of requirement dependence extraction and expert extraction are compared and analyzed. The results show that the proposed method has high feasibility and accuracy which can save a lot of human resources and cost.