Predicting Credit Card Approvals

Build a machine learning model to predict if a credit card application will get approved

Project Tasks

1. Credit card applications

Load the Dataset via pandas

2. Inspecting the applications

Check the various columns and their datatypes and trends

3. Handling the missing values (part i)

Using NumPy for dealing with missing values

4. Handling the missing values (part ii)

Dealing with NULLS

5. Handling the missing values (part iii)

Dealing with missing numerics and filling it with values

6. Preprocessing the data (part i)

Using sklearn's preprocessing Label Encoders for converting non-numeric values to numeric for faster calculations

7. Splitting the dataset into train and test sets

Using sklearn's Model selection for splitting dataset into train and test set

8. Preprocessing the data (part ii)

Using sklearn's MinMaxScaler for normalizing the dataset for faster computation

9. Fitting a logistic regression model to the train set

Using sklearn's Logistic Regression for prediction model

10. Making predictions and evaluating performance

Using confusion matrix to view our model's performance

11. Grid searching and making the model perform better

Using GridSearchCV with multiple variable parameter for better model

12. Finding the best performing model

Displaying the best parameter and its accuracy after gridsearch cv works into our test set