A Tool for Automatically Identifying Semantic Conflicts in User Stories by Combining NLP and BERT Model

In agile software development, user stories are provided by different stakeholders, and there may be semantic conflicts between them. It takes a lot of time and manpower to identify semantic conflicts from a large number of user stories, and the identification effect is not ideal. This paper proposes a semantic conflict identifying tool. The tool mainly includes two parts: one is to identify whether there are syntax problems or ambiguity problems in each user story, and the other is to identify whether there are semantic problems between user stories. The tool combines the NLP and BERT model for improving the syntactic and semantic quality of user stories through feedback and interaction. Our preliminary results on 9060 user stories show an F1-Score of 92%. A demo video of this tool is available at https://youtu.be/EGYyI9RclRc.