Technical screening – model development discussion

Describe at least two methods to handle missing values in a training set. What are the pros and cons of each? Give two strategies for treating outliers and explain when you would prefer each. Which metrics would you use to evaluate classification and regression models? Justify your choices. Pick one machine-learning algorithm of your choice and walk us through how it works step by step. List the key hyperparameters in XGBoost and explain their impact on the model.

Cover imputation, deletion, winsorization, MSE/ROC/AUC/F1, algorithm mechanics, learning_rate, max_depth, n_estimators …

Onemain financial Data Scientist interview question: Handle Missing Values and Outliers in Machine Learning. {"blocks": [{"key": "c2fa12f7", "text": "Scenario", "type": "header-two", "depth": 0, "inlineStyleRanges": [], "entityRanges": [], "data": {...