---

Loan Approval Prediction Project

Overview

This project focuses on predicting loan approvals using a dataset with various applicant details. The process includes data preprocessing, feature engineering, model training, and evaluation using Decision Tree and Naive Bayes classifiers.

Project Structure

• loan-train.csv: Training dataset
• loan-test.csv: Test dataset
• loan_approval_prediction.ipynb: Jupyter notebook with detailed analysis and code

Skills and Technologies Used

• Data Preprocessing
  • Handling missing values
  • Data normalization (log transformation)
• Feature Engineering
  • Categorical data encoding
  • Feature scaling
• Model Training
  • Decision Tree Classifier
  • Naive Bayes Classifier
• Model Evaluation
  • Accuracy calculation
  • Performance metrics analysis
• Data Manipulation
  • pandas for data manipulation
  • numpy for numerical operations
• Machine Learning
  • sklearn for model training, evaluation, and preprocessing
• Data Visualization
  • Matplotlib for histogram plotting of transformed features

Installation

1. Clone the repository:

   git clone https://github.com/crazyNerrd/Loan-Approval.git

2. Install the required packages:

   pip install -r requirements.txt

Usage

1. Navigate to the project directory:

   cd loan-approval-prediction

2. Open the Jupyter notebook:

   jupyter notebook loan_approval_prediction.ipynb

Data Preprocessing

• Handle missing values using mode and mean imputation.
• Apply log transformation to LoanAmount and TotalIncome for normalization.
• Encode categorical variables using LabelEncoder.
• Scale features using StandardScaler.

Model Training and Evaluation

• Split the dataset into training and testing sets using train_test_split.
• Train Decision Tree and Naive Bayes classifiers.
• Evaluate models' accuracy using metrics from sklearn.

Test Data Preparation

• Apply the same preprocessing steps to the test dataset.
• Make predictions using the trained Naive Bayes classifier.

Results

• Display accuracy scores for both Decision Tree and Naive Bayes classifiers.
• Provide insights on model performance and potential improvements.