Optimizing Software Bug Prediction with Explainable AI: ASHAP-Based Analysis Using Random Forests
Students & Supervisors
Student Authors
Supervisors
Abstract
Softwarebugpredictionmodelshelpoptimize testingresourcesandreducede velopmentcosts.Traditionalbugdetectionmethodsrelyonpredefinedrulesor machinelearningmodels,whichoftenlacktransparency,makingdebuggingdifficult. Whilemachinelearningmodelsachievehighaccuracy, theirblack-boxnaturelimits interoperability,makingitchallengingfordeveloperstounderstandthepredictions. Toaddressthis issue, thisstudyproposestheuseof ((XAI)withShapleyAddi tiveexPlanations(SHAP)toenhancebugdetectionbyprovidingclear insights into modelpredictions.SHAPispreferredoverLIMEbecauseof itsconsistentandaccu ratefeatureattribution,makingitmorereliablefordebugging.Additionally,XAI is chosenovertraditionalrule-basedordeeplearning(CNN)modelsbecauseitimproves transparency,allowingdeveloperstoquicklyidentifyandresolvesoftwaredefects.In thisstudy,weusedSHAPforinteroperabilityandtheNASAPROMISEdataset.The experimentalresultsshowthatourSHAP-basedmodelachieved85%accuracy.This approachaims tobridge thegapbetweenAI-drivenbugdetectionanddeveloper friendlyinterpretability,ensuringmoreefficientandtrustworthydebugging
Keywords
Publication Details
- Type of Publication:
- Conference Name: 3rd International Conference on Big Data, IoT and Machine Learning (BIM 2025)
- Date of Conference: 25/09/2025 - 25/09/2025
- Venue: Dhaka International University, Bangladesh