Semantically Enhanced Software Traceability and its Applications
Jin Guo - PhD Candidate, Computer Science and Engineering, University of Notre Dame
Jan. 11, 2017, 10 a.m. - Jan. 11, 2017, 11 a.m.
MC437
Modern Software and Systems engineering projects produce large quantities of data (artifacts) as a natural byproduct of the engineering process. These artifacts, including user stories, requirements, design documents, source code, and test cases, are inter-connected, and their connectivity can be used to answer questions of both the software product and its development process. The connections between software artifacts are represented as trace links. For safety-critical systems the need for creating and maintaining trace links is prescribed by certifying bodies in order to demonstrate the systems are safe for use.
Creating trace links manually is time consuming and error prone and several studies have shown that it produces incomplete, redundant, and inaccurate links. Automated solutions use information retrieval and machine learning techniques to generate trace links; however, current techniques fail to understand the rich semantics in software artifacts or to integrate domain knowledge into the tracing process and therefore tend to deliver imprecise and inaccurate results. In this talk, I will explain how to achieve more accurate trace link generation through extracting and leveraging embedded semantic information from software artifacts written in natural language. I will draw examples from specific projects involving discovery of domain-concepts, heuristic based reasoning over these concepts, and deep learning. I will also illustrate how trace links can deliver actionable intelligence in support of software engineering activities such as requirements satisfaction analysis, safety certification, and project Q&A.
Jin Guo is a PhD Candidate in Computer Science and Engineering at the University of Notre Dame. Her PhD work, supported by the US Natural Science Foundation, focused on semantically enhanced software trace link generation for safety critical systems. Her broad research interests include Software Traceability, Requirements Satisfaction Analysis, Natural Language Processing, and Neural Networks. Before she started her PhD, Jin Guo was a researcher in the Communication Technology Lab at Fuji Xerox Co., ltd. in Japan. For more information about Jin Guo’s work, please visit her website at http://jguo-web.com/.
All Are Welcome
