├── Credit Risk Modelling in Python.ipynb
├── README.md
└── UCI_Credit_Card.csv


/README.md:
--------------------------------------------------------------------------------
 1 | # Credit-Risk-Modelling-in-Python 
 2 | Credit risk refers to the chance that a borrower will be unable to make their payments on time and default on their debt. It refers to the risk that a lender may not receive their interest due or the principal lent on time.
 3 |  
 4 | ## More detail explanation
 5 | Read in my blogs https://medium.com/mlearning-ai/credit-risk-modelling-in-python-7b21a0b794b1 
 6 | 
 7 | ## Objective
 8 | The objective of this project is to predict the probability of default on a given obligation, in this case, credit cards. This will allow the generation of strategies that minimize the risk of deterioration of the client's financial health.
 9 | 
10 | ## Approach
11 |  1. Data preparation and Pre-processing
12 |      - Checking missing values
13 |      - Outliers removal
14 |      - Encoding concept
15 |  2. Feature Engineering and Selection
16 |      - features selection
17 |      - Scaling the features
18 |      - Balancing the data
19 |  3. Model Development and Model Evaluation
20 |      - Logistic regression
21 |      - Random Forest
22 |      - Xgboost
23 |      - Hyperparameter tuning
24 |      - Cross validation
25 | 
26 | ### Dependencies
27 |  
28 | * numpy
29 | * matplotlib
30 | * scikit-learn
31 | * XGBOOST
32 | * seaborn
33 | * pyhthon 3.9
34 | 
35 | ## Usage
36 | 
37 | Just run `jupyter notebook` in terminal and it will run in your browser.
38 | 
39 | - Install Jupyter [here](http://jupyter.readthedocs.io/en/latest/install.html) i've you haven't.
40 | 
41 | - install xgboost by using `pip install xgb` in command line prompt/ anconda i've you haven't.
42 | 
43 | ## Machine learning algorithm
44 | - Logistic regression
45 | - Random Forest
46 | - Xgboost Classifer
47 | - Cross validation
48 | 
49 | ## Model Evaluation
50 | - we have acheived the final model accuracy 82%
51 | 
52 | ## Steps to run this model in your system 
53 | - Clone or download the repo.
54 | - Open command prompt in the downloaded folder
55 | 
56 | ## Conclusion:
57 | We have built the credit risk model by using python. we have tried with different machine learning algorithms ie. Logistic regression, Random Forest, and Xgboost classifier. we have also done hyperparameter tuning and cross-validation. The final accuracy of the model we achieved was 82%.
58 | 
59 | ## Dataset:
60 | - https://www.kaggle.com/uciml/default-of-credit-card-clients-dataset 
61 | 
62 | 


--------------------------------------------------------------------------------