I am trying to build a simple anomaly detection solution.

I have a dataset with normal observations. It has about a 100 values. I calculate the np.percentile(dataset1) of this dataset.

I have another dataset of new observations which contains about 10 values. I calculate the np.percentile(dataset2) of this dataset as well.

Is it fair to say if the 25th, 50th and 75th percentile of dataset2 are 25%(can be any number based on the level of sensitivity you desire) or more different than dataset1, then the observations in dataset2 are anomalous since dataset1 is suppose to be a dataset of normal observations?


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.