"Accuracy and binominal classification"

UsernameUsername Member Posts: 39 Maven
edited June 2019 in Help
Hi,

the operator for evaluating binominal classification tasks does not offer accuracy. I guess this is due to the fact that accuracy doesn't make sense for binominal classification but I haven't figured out so far why this should be the case. I hope somebody has an explanation.

Thanks
Tagged:

Answers

  • IngoRMIngoRM Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, Community Manager, RMResearcher, Member, University Professor Posts: 1,751 RM Founder
    Hi,

    accuracy of course also makes sense for binominal classification tasks. But it one of those measure which works for all types of classification and hence can be found in the (polynominal and hence more generic) ClassificationPerformance operator. You can calculate both accuracy and things like precision, recall (and other binominal measures) by performing both operators on the labeled example set (don't forget to activate "keep_example_set" for the first one).

    Cheers,
    Ingo
Sign In or Register to comment.