Mining explicit rules for software process evaluation

Abstract

We present an approach to automatically discovering explicit rules for software process evaluation from evaluation histories. Each rule is a conjunction of a subset of attributes in a process execution, characterizing why the execution is normal or anomalous. The discovered rules can be used for stakeholder as expertise to avoid mistakes in the future, thus improving software process quality; it can also be used to compose a classifier to automatically evaluate future process execution. We formulate this problem as a contrasting itemset mining task, and employ the branch-and-bound technique to speed up mining by pruning search space. We have applied the proposed approach to four real industrial projects in a commercial bank. Our empirical studies show that the discovered rules can precisely pinpoint the cause of all anomalous executions, and the classifier built on the rules is able to accurately classify unknown process executions into the normal or anomalous class.

On the Validation of a Specific Development Process for Scientific Software using the Inspection Technique DOI - 10.5752/P.2316-9451.2015v3n2p3

“…In this scenery, it is possible to make the simulation and verify how the process treats a particular situation (NAVARRO; HOEK, 2005). Sun, Du and Chen (2013) presented an approach to automatically discover explicit rules for software process evaluation from the evaluation histories. Each rule is a conjunction of a subset of attributes in a process execution, characterizing the reason why the execution is normal or anomalous.…”

“…Each rule is a conjunction of a subset of attributes in a process execution, characterizing the reason why the execution is normal or anomalous. Sun, Du and Chen (2013) formulated this problem as a contrasting item set mining task and employ the branch-and-bound technique to speed up mining by pruning search space. Ferreira and Moita (2010) also suggested an approach to compare the formal model and the execution of the process, but they adopted concepts based on software inspection.…”

Mining succinct predicated bug signatures

“…Yan et al mine discriminative subgraph patterns via leap search [27]. Sun et al mine contrasting patterns for software process evaluation [25].…”

Querying sequential software engineering data

“…These violation patterns are learnt from historical process execution data using a contrasting pattern mining algorithm [26]. They are meaningful to the bug management process of a specific commercial bank introduced in [14,13].…”

“…Their approach first learns a binary classification model from the historical evaluation results, and then applies the model to classify future process executions. Sun et al [26] further improves this approach by mining explicit evaluation rules from evaluation history. Each rule is capable of explaining why a process execution is evaluated as normal or anomalous.…”