Improving Fault Localization by Integrating Value and Predicate Based Causal Inference Techniques

This is the artifact for the paper "Improving Fault Localization by Integrating Value and Predicate Based Causal Inference" (ICSE 2021). This artifact is an Oracle Virtual Box (Virtual Machine) image that can be downloaded and imported as an appliance using Oracle’s Virtual Box software (https://www.virtualbox.org/). We assumed no advanced technology skills other than downloading the Virtual Box software and the Virtual Machine (VM) imagine and clicking on “Import Appliance” to be able to get the VM image set-up. Running our experiments is as simple as running a command for a bash script (e.g. ~./UniVal.sh Closure 62)
This artifact facilitates the use of UniVal for localizing and debugging defects from the Defects4J (v 2.0) suite. It implements the algorithm presented in the paper. In addition, this artifact contains the implementation of our prototype tools Predicate_Transformer and GSA_Gen mentioned in the paper. Finally, this artifact contains our implementations of the competing techniques namely, NUMFL-DLN/PLY, Elastic Predicates (ESP), Baah2010, Ochiai, D-Star and Predicate Switching fault localization suspiciousness metrics.