A Proposal to Combine Probabilistic Reasoning with Case-Based Retrieval for Software Troubleshooting

Akihiro Shinmori

Analysis of a real case of software troubleshooting made it clear that probabilistic models, or Bayesian networks are suitable for modeling software troubleshooting processes in today’s computing environments. As a way to support software troubleshooting by reusing skillful support engineers’ expertise, I propose to use Bayesian networks and probabilistic reasoning for case-based retrieval. The outline of query reformulation using Bayesian network is presented.


This page is copyrighted by AAAI. All rights reserved. Your use of this site constitutes acceptance of all of AAAI's terms and conditions and privacy policy.