Confusion matrices, Classification report provide insights in terms of accuracy. how to evaluate classification models on tasks while tackling domain specfic problems e.g mechanical systems engineering problems, where we are mostly dealing with time-series data or its frequency domain features? and want to use classification models for tasks such as Anomaly detection, fault classification? which types of visulization will explain my research more accurately?
Maybe it helps you to get an approach:
https://medium.com/eatpredlove/time-series-cross-validation-a-walk-forward-approach-in-python-8534dd1db51a
Classification models can be evaluated with F1_score, exists some libreries in sci-kit learn, this library is used in python, and for their respective visualization depends on the technique that you uses, for example, if you are using logistic regression is different that if you are using a k-mean algorithm. If you want I can send you a script made in Python with a logistic regression example.Retreats
- Badges
- Science topic
- Similar topics
- Data Mining
- Classification