Software Requirement-Related Information Extraction from Online News using Domain Specificity for Requirements Elicitation
Mutia Rahmi Dewi, Indra Kharisma Raharjana, Daniel Fernando Siahaan, Chastine Fatichah · 2021
Stakeholder involvement is vital in the requirements elicitation process. However, a lack of stakeholder involvement in the project often occurs. Various data sources, such as business processes and software documentation, will help determine what stakeholders need. The limitation of these documents is that they are technical and do not contain numerous stakeholder's needs. This paper aims to extract information related to software requirements from online news using domain specificity. The online news comprises elements that compose the user story, namely: the aspect of who (stakeholder), the aspect of what (functional), and the aspect of why (reason). User stories are standard practices used to capture and write functional software requirements in agile software development. This paper proposes to use domain specificity to extract aspects of what from online news to increase understanding of domain knowledge, especially software functionality. We use SRS documents and online news as data sources. Based on these datasets, domain specificity is calculated to produce software specific vocabulary. POS tags are being used to extract software requirements-related information from online news. When examining two datasets, our approach improved precision and recall with average values of 0.56 and 0.579.