How should the Specialist frame this business problem?
Streaming classification
Binary classification
Multi-category classification
Regression classification
Explanations:
While streaming classification involves processing data in real time, it does not specifically indicate the nature of the prediction being made. The primary goal is to classify transactions as fraudulent or not, which is a binary classification task.
This is the correct framing of the problem, as it involves predicting whether a transaction is fraudulent (1) or not (0). The output is a binary decision, making it a binary classification problem that returns the probability of fraud.
Multi-category classification involves classifying instances into more than two categories. In this case, there are only two possible outcomes (fraudulent or not fraudulent), so this framing is not applicable.
Regression classification is not a standard term. Regression typically deals with predicting continuous values rather than classifying instances into categories. This problem specifically requires classifying transactions, not predicting a numerical value.