Oluwakemi-moyin/Loan-Prediction-Model---Africa-Agility-

# Loan-Prediction-Model---Africa-Agility Tools used: Python ## Introduction The purpose of a loan prediction model is to assess the likelihood of a loan applicant repaying a loan, thereby aiding in risk assessment and loan approval decisions. ## Objective The goal of a loan prediction data analysis is to: - Develop a model that accurately predicts the likelihood of loan repayment (or default). - Identify key factors (features) that influence loan approval and repayment. - Provide insights to improve loan risk assessment and decision-making processes. - Minimize financial risk for lenders by reducing loan defaults. - Potentially streamline the loan approval process. ### The data source used: LOAN PREDICTION DATASET 2. Data Summary & Insights 2.1 Data Overview: - The dataset consisted of 614 Loan records and 13 columns. - Key attributes including Loan_ID, Gender, ApplicantIncome, CoapplicantIncome, LoanAmount, Credit_History, Property_Area, and Loan_Status. - The data showed a wide distribution of Loan applicants across different distinct Property areas, income ranges and genders (male and female). 2.2 Key Observations: ### Findings: - The analysis revealed that credit history, applicant income, and loan amount are the strongest predictors of loan approval. - Random Forest Algorithm result: The model achieved an accuracy of 78% on the test dataset - Logistic Regression Algorithm result: The model achieved an accuracy of 79% on the test dataset - Support Vector Machines (SVMs) result: The model achieved an accuracy of 79% on the test dataset ### Patterns observed in loan data: - Credit_History is the most important feature. A credit history indicates how reliably a person has repaid debts in the past. - LoanAmount :The amount of the loan is also a very strong predictor. Larger loans may be riskier and are thus more closely scrutinized. - ApplicantIncome :The applicant's income is a crucial factor in determining their ability to repay a loan. Higher i …