Semantic Analysis of Functional and Non-Functional Requirements in Software Requirements Specifications

Software Requirements Specifications (SRS) documents are important artifacts in the software industry. A SRS contains all the requirements specifications for a software system, either as functional requirements (FR) or non-functional requirements (NFR). FRs are the features of the system-to-be, whereas NFRs define its quality attributes. NFRs impact the system as a whole and interact both with each other and with the functional requirements. SRS documents are typically written in informal natural language [1], which impedes their automated analysis. The goal of this work is to support software engineers with semantic analysis methods that can automatically extract and analyze requirements written in natural language texts, in order to (i) make SRS documents machine-processable by transforming them into an ontological representation; (ii) apply quality assurance (QA) methods on the extracted requirements, in order to detect defects, like ambiguities or omissions; and (iii) attempt to build traceability links between NFRs and the FRs impacted by them, in order to aid effort estimation models.