https://github.com/arpanpramanik2003/loan-status-prediction
The **Loan Status Prediction Model** predicts loan approval based on applicant details like income, credit history, and loan amount. It uses data preprocessing, an SVC model, and achieves around 79% accuracy. The trained model is saved for future use.
https://github.com/arpanpramanik2003/loan-status-prediction
data-processing data-science loan-approval-prediction loanprediction machine-learning mlpipelines model-evaluation pickle predictive-analytics sklearn svm-classifier
Last synced: 7 months ago
JSON representation
The **Loan Status Prediction Model** predicts loan approval based on applicant details like income, credit history, and loan amount. It uses data preprocessing, an SVC model, and achieves around 79% accuracy. The trained model is saved for future use.
- Host: GitHub
- URL: https://github.com/arpanpramanik2003/loan-status-prediction
- Owner: arpanpramanik2003
- License: apache-2.0
- Created: 2024-12-20T14:28:34.000Z (10 months ago)
- Default Branch: master
- Last Pushed: 2025-01-22T17:50:00.000Z (8 months ago)
- Last Synced: 2025-01-22T18:36:43.086Z (8 months ago)
- Topics: data-processing, data-science, loan-approval-prediction, loanprediction, machine-learning, mlpipelines, model-evaluation, pickle, predictive-analytics, sklearn, svm-classifier
- Language: Jupyter Notebook
- Homepage: https://loan-status-prediction-2upm.onrender.com
- Size: 43 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Loan Status Prediction Model
## Project Overview
This project focuses on predicting loan approval status using machine learning techniques. The dataset contains various features related to loan applicants, and the goal is to classify whether a loan will be approved or not.
## Dataset
- The dataset used is `Loan_status.csv`.
- It consists of 614 rows and 13 columns.
- Key features include applicant's income, loan amount, credit history, and property area.
## Data Preprocessing
1. Checked for null values and removed rows with missing data.
2. Converted categorical values into numerical values:
- Loan_Status: 'Y' to 1, 'N' to 0
- Dependents: '3+' to 4
3. Split the data into features (X) and target (Y).
4. Defined categorical and numerical columns for preprocessing.
## Model Training
1. Used `StandardScaler` for numerical feature scaling.
2. Applied `OneHotEncoder` for categorical feature encoding.
3. Built an SVM model with a linear kernel.
4. Created a pipeline for data preprocessing and model training.
## Model Evaluation
- **Training Accuracy:** 81.02%
- **Testing Accuracy:** 79.17%
## Model Deployment
- The trained model is saved as `loan_status_model.pkl` using the `pickle` module.
## Libraries Used
- pandas
- numpy
- matplotlib
- seaborn
- scikit-learn
- pickle
## Usage
1. Load the trained model from `loan_status_model.pkl`.
2. Provide applicant data for prediction.
3. Get loan approval status (1 for approved, 0 for not approved).
## Conclusion
This loan status prediction model can help financial institutions automate loan approval decisions based on applicant details, improving efficiency and accuracy.