I have a real-time scenario, finding out whether a transaction is fraudulent or not. I have a dataset that contains only fraudulent transactions. For any binary classification algorithm, we may need both classes. How do you handle this type of situation in machine learning?

  • 1
    $\begingroup$ see for example one class svm, isolation forest, autoencoders, .. $\endgroup$ – Nikos M. Jun 13 at 4:44
  • $\begingroup$ Welcome to DatascienceSE. You need to treat this problem as binary classification problem with two classes as Fraud and Not Fraud. Yes, for any binary classification algorithms, we need both classes. $\endgroup$ – Devashish Prasad Jun 13 at 6:45

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.