Dataset Open Access

Replication package of "Good Things Come In Threes: Improving Search-based Crash Reproduction With Helper Objectives"

Derakhshanfar, Pouria; Devroey, Xavier; Zaidman, Andy; van Deursen, Arie; Panichella, Annibale

The replication package for the study about using new helper objectives (MOHO) for crash reproduction. This study has been accepted at ASE 2020.

 

Abstract:

Evolutionary intelligence approaches have been successfully applied to assist developers during debugging by generating a test case reproducing reported crashes. These approaches use a single fitness function called Crash Distance to guide the search process toward reproducing a target crash. Despite the reported achievements, these approaches do not always successfully reproduce some crashes due to a lack of test diversity (premature convergence). In this study, we introduce a new approach, called MO-HO, that addresses this issue via multi-objectivization. In particular, we introduce two new Helper-Objectives for crash reproduction, namely test length (to minimize) and method sequence diversity (to maximize), in addition to Crash Distance.

We assessed MO-HO using five multi-objective evolutionary algorithms (NSGA-II, SPEA2, PESA-II, MOEA/D, FEMO) on 124 hard-to-reproduce crashes stemming from open-source projects. Our results indicate that SPEA2 is the best-performing multi-objective algorithm for MO-HO.

We evaluated this best-performing algorithm for MO-HO against the state-of-the-art: single-objective approach (Single-Objective Search) and decomposition-based multi-objectivization approach (De-MO). Our results show that MO-HO reproduces five crashes that cannot be reproduced by the current state-of-the-art. Besides, MO-HO improves the effectiveness (+10% and +8% in reproduction ratio) and the efficiency in 34.6% and 36% of crashes (i.e., significantly lower running time) compared to Single-Objective Search and De-MO, respectively. For some crashes, the improvements are very large, being up to +93.3% for reproduction ratio and -92% for the required running time. 

Files (8.1 GB)
Name Size
STAMP-project/Botsing-multi-objectivization-using-helper-objectives-application-1.0.zip
md5:8b13928c738cabe57b5992ca6d8defc6
8.1 GB Download
77
31
views
downloads
All versions This version
Views 7777
Downloads 3131
Data volume 250.2 GB250.2 GB
Unique views 6767
Unique downloads 2626

Share

Cite as