I know case-based reasoning has four stages: retrieve, retain, re-use and revise.

Used for solving new problems by adapting solutions that were used to solve old problems, like car issues.

The advantages of it are that solutions are proposed quickly and there's no need to start from scratch, but is there a real-world problem where using this would not be suitable?

Just trying to better my understanding of it.

  • $\begingroup$ Hi and welcome to this community. What does case-based reasoning have to do with reinforcement learning (given that you added this tag to your question)? $\endgroup$ – nbro May 13 '19 at 17:42
  • $\begingroup$ Wasn't sure what to put as the tag since case-based reasoning isn't an option, apologies, what should I change it to? $\endgroup$ – l15 May 13 '19 at 18:28
  • $\begingroup$ Which source have you read that mentions case-based reasoning in the context of AI? $\endgroup$ – nbro May 13 '19 at 18:38
  • $\begingroup$ Well I'm learning it as part of a module for AI at uni and the way we're being taught is how it's used to compare old solutions with new problems. Using it with applications to find cars with similar features etc. The reason I'm asking about when it can't be used is I found a question on it on a past paper and when I asked my lecturer if he could provide extra information as the notes don't explain when not to use it, I just got told no. $\endgroup$ – l15 May 13 '19 at 18:51
  • $\begingroup$ Have a look at this video Case-Based Reasoning - AI 101. $\endgroup$ – nbro May 13 '19 at 22:19

Your Answer

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

Browse other questions tagged or ask your own question.