Risk correlation is an important topic in software development. This study applied meta-analysis to both collect risk checklists and compose a natural language description set of 1932 sentences. Co-word analysis was performed to obtain a word co-occurrence matrix. The correlation network features were then analyzed in the form of a social network. Characters including density, centrality, distance and core/periphery were put forward about the risk correlation network. The high frequency risk word list and related network conclusions are helpful for software risk management.
Published in:
Business Intelligence and Financial Engineering (BIFE), 2012 Fifth International Conference on
Date of Conference: 18-21 Aug. 2012