Change-Patterns Mapping: A Boosting Way for Change Impact Analysis

Change impact analysis (CIA) is a specialized process of program comprehension that investigates the ripple effects code change in software system. In this paper, we present boosting way for via mapping historical change-patterns to current CIA task cross-project scenario. The reflect coupling dependencies between changed entities set. A traditional tool (such as ImpactMiner) outputs an initial set starting entity. To boost tool, our approach retrieves equivalent entity from various sets Then, and rest are mapped at hand. For task, if involves similar change-pattern with when comparing change-pattern, will reward impacted confidence Accuracy improvements observed experiments applying method three famous tools, i.e., ImpactMiner, JRipples ROSE.