期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2017
卷号：95
期号：5
出版社：Journal of Theoretical and Applied
摘要：Use case modeling is an important requirements engineering technique which plays an important role in describing the systems specifications and facilitating systems development. The use of linguistic representations of system requirements as a source of information for generating use case models is a challenging task and can be considered relatively a new field. This paper has tackled the problem of extracting the required elements that are needed to automatically generate use case diagrams from specification documents which are written in common natural language. Therefore, we have developed an automated system which employs the Natural Language Processing (NLP) techniques to parse specifications syntactically based on a predefined set of heuristic rules. Furthermore, our system incorporates the capability of analyzing and understanding the English text as a semantic unit to infer some important linguistic features such as reference, comparing and additive cohesive devices. The extracted information is then mapped into actors and use cases, which are the basic elements of use case diagrams. Our proposed approach was evaluated using both recall and precision performance measurements. The experiments revealed that our system has an average of 96% recall and 84% precision.
关键词：Use case diagrams; Natural Language Processing; User requirements analysis; Automatic Diagrams Generation; Information Extraction.
