, or statistics.
2. Be careful with those missing data. They can be empty string, -1, -98, -99, etc. You will need to check the data and var dictionary to make your best judgement.
2. Use those selected columns to predict the "loan_default" column. You will try 3 machine learning algorithms
* Logistic regression
* Naive Bayes Classifier
3. For each algorithm, you should select features and fit the model, then predict and evaluate.
4. Try different techniques to improve the model score. You can choose different columns, transform data and normalize data. Show your improvements.