Pls refer screenshot for sample data. As can be seen most of the fields in data are textual and highly correlated but each row has unique values and hence won't be right to call it categorical. I tried to break down column 1 in two tokens to ABC and X1 and tried KModes clustering where number of clusters = unique values in column1. However each cluster do not have equal density hence some bad data with high density is classified as normal and good data with low density is marked anomalous. I want to have unsupervised algo where I can force it to use Key1+Key2 as composite primary key for clustering. Algo should then ensure data follows some pattern in each cluster e.g. (alias1, alias2 moves in unison. Multipliers are in close range.) Kindly suggest what would be the best anomaly detection algo and how to approach this problem. Sample Data



Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.