Implementation-Centric Classification of Business Rules from Documents

摘要: In large multi-site multi-vendor projects, studying requirement documents to understand the problem domain and inferring possible solution posed is an important activity in Requirements Engineering. The process of reading User require-ments Specification (URS) create Software Requirement Speci-fication (SRS) a knowledge intensive that precedes sev-eral other Engineering (SE) activities such as design test plans. Automated Interpretation URS terms implementation-specific elements for software engineers' consumption has been reported past. aim interpretation reduce effort associated with manual extraction subsequently, their "translation" into primitives understood by those who must build intended software. this paper, we present deep learning model implementation-centric classification one element, namely, business rules. We discuss approach based on Bidirectional Long Short Term Memory Network (BiLSTM) capture context information each word, followed attention aggregate useful infor-mation from these words get final classification. Our adopts end-to-end architecture does not rely any handcrafted features.