1. Machine Learning Model Design for ATO Detection
Design a machine learning model for account takeover (ATO) detection, covering the entire process from model building to deployment. Address potential issues such as database update latency.
2. Debugging Machine Learning Code
Given a machine learning codebase that you have cloned from a git repository to your local environment, identify and debug errors. The errors include a range error in a for loop, a mix-up between 'predict' and 'predict_proba' functions, and a definition error regarding the AUC score. Explain your debugging process for each issue.
3. ML System Design for Account Takeover Prevention
Design a machine learning system for preventing account takeover, similar to a fraud detection system. Discuss the definition of account takeover, business requirements/success metrics, ML model options, potential features, system design diagram, database options, model serving options, and how to handle large traffic (considering load balancers, caching, etc.).
4. ML Integration: Binary Classification
Perform a binary classification task using a dataframe that contains both numerical and categorical data. Describe your approach to integrating machine learning into this problem.
