Skip to content

Analyze data & make machine learning model with binary classification method based on the credit history of the debtors.

Notifications You must be signed in to change notification settings

MuharomRahZ/ml-binary-classification-bad-debt

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 

Repository files navigation

Bad Debt/Credit Scoring Machine Learning Model - Binary Classification (CAPSTONE PROJECT)

This capstone project is a final work/project from Junior Data Science program (intermediate course) in an online summer class with RMDS Narasio Data (https://www.instagram.com/narasiodata/). In this work I have multiple objectives to be achieved:

1. Perform Bad Debt / Credit Analysis

First, I was asked to analyze the dataset from a 'bank' to get insights from debtors debt/credit history. Then, I was asked to do an assessment of the data & determine which debtors have good credit scores and which debtors have bad credit scores, based on the credit history of the debtors.

2. Build Machine Learning Model (Binary Classification)

Then, the bank asked me to create an automatic creditworthiness assessment system, based on the credit history data of their debtors. In this work, I was asked to create a machine learning model using the binary classification method. The machine learning model created based on this method will later produce a bad credit scoring prediction based on automatic classification of the data performed by the machine learning model.

3. Predict New Data

After the machine learning model is successfully created, this machine learning model will be applied to the new data & will generate a bad credit score based on the binary classification method in the machine learning model.

4. Give Feedback Recommendation

Lastly, I was asked to provide feedback in the form of conclusions & recommendations on the results of the machine learning model that have been made, and business decisions that the bank can make or implement in the future.

Releases

No releases published

Packages