I've been struggling with the connection between knowledge based AI systems and Bayesian inference for a while now. While I continue to sweep through the literature, I would be happy if someone can answer these questions directly -

  1. Are Bayesian inference based methods used in reasoning or Q/A systems -- to arrive at conclusions about questions whose answers are not directly present in the knowledge base?
  2. In other words, if a Q/A system doesn't find an answer in a Knowledge base, can it use Bayesian inference to use the available facts to suggest answers with varying likelihoods?
  3. If yes, could you point me to some implementations?

Yes, it is possible to combine probabilistic / bayesian reasoning and a traditional "knowledgebase". And some work along those lines has been done. See, for example, ProbLog ("Probabilistic Prolog") which combines logic programming and probabilistic elements. See:


Another project to look at is Pr-OWL ("Probabilistic OWL") which adds Bayesian reasoning to the Semantic Web stack.

Of course neither of these deals specifically with QA systems, but both represent some work on at least the foundational aspect of combining traditional logic and/or ontologies, with probabilistic approaches. Building a QA system on top of that is an exercise for the reader...

| improve this answer | |
  • $\begingroup$ Your answer led me to some very useful links. I'm still exploring the plethora of new things I hadn't seen before. Thank you so much for that. Would come back if I find some insight worth sharing. Till then, at least, this stands as the accepted correct answer $\endgroup$ – PintoUbuntu Jan 21 '17 at 9:22

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.