Data-mining in Support of Detecting Class Co-evolution.

摘要: In an evolving system maintained over a long time period, there exist many non-trivial relationships among classes, such as class co-evolutions, which usually are not easily perceivable in the source code. However, unfortunately, continuing evolution of large, long-lived systems leads to lost information about these hidden relationships. this paper, we propose method for recovering knowledge by data mining method. This relies on UMLDiff algorithm that, given sequence UML models system, surfaces design-level changes its life span, thus eliminating need high quality modification reports and nonintuitive software code-based metrics. We employ Apriori association rule transactional database modifications, elicit previously unknown or undocumented co-evolving relations two more classes. The recovered facilitates overall understanding planning future maintaining activities. report one real world case study evaluating our approach.