Multivariate anomaly detection

[LR_9796586]

I'm working on a anomaly detection dataset for which I don't have labels for the target class, hence it is an unsupervised machine learning problem. As I understand it, the anomaly detection module  is univariate, is there another available tool in ThingWorx to approach this kind of problems? such as isolation forest or autoencoders?

[nsampat · Accepted solution]

@LR_9796586 ,

Thank you for posting your question to the PTC Community.

Regarding the use of Unsupervised Learning, this is not a feature of ThingWorx Analytics at this time.

The methods we do support can be found on our HelpCenter Topics:

ThingWorx Analytics 8.x - Learners and Ensemble Techniques
ThingWorx Analytics 9.x - Learners and Ensemble Techniques

Please let me know if this answers your question by marking this topic "Accepted Solution"

Regards,

Neel

[slangley]

Hi @LR_9796586

If the previous response was helpful, please mark it as the Accepted Solution for the benefit of others with the same questions.  If you still have questions, please let us know.

Regards.

--Sharon