1
$\begingroup$

I was having a discussion with a coworker with regard to machine learning classifiers--specifically we were thinking about Random Forest but that is somewhat besides the point. The use case we were considering was one where we want to model the behavior of a system operating under normal conditions and have a classifier that could identify when the system is in an anomalous state based on the features we selected. I mentioned the difficulty of having an adequate dataset to train that includes a significant amount of data for the system when its in an anomalous state, but my coworker countered that he believed you could train using just data for the system under normal conditions and that would be good enough.

That seems to strike a sour note in my head, so I was hoping to get someone else's opinion as my understanding of ML concepts is pretty limited.

$\endgroup$
0
$\begingroup$

For anyone else interested in this, I found out that such a concept does exist and has been researched to some extent. You can find information and research on the topic by searching for One-class classifiers or unary classifiers.

$\endgroup$

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.