Create loanprediction.py

NDV_Code_By_PoojithB/loanprediction.py

@@ -0,0 +1,75 @@
+# Loan Dataset Analysis
+import pandas as pd
+import numpy as np
+import seaborn as sns
+import matplotlib.pyplot as plt
+from scipy.stats import ttest_ind
+
+# Load dataset
+loan_path = 'loan_data.csv'
+loan_df = pd.read_csv(loan_path)
+
+# Handle missing values
+loan_df['Gender'].fillna(loan_df['Gender'].mode()[0], inplace=True)
+loan_df['Married'].fillna(loan_df['Married'].mode()[0], inplace=True)
+loan_df['Dependents'].fillna('0', inplace=True)
+loan_df['Self_Employed'].fillna(loan_df['Self_Employed'].mode()[0], inplace=True)
+loan_df['Credit_History'].fillna(loan_df['Credit_History'].mode()[0], inplace=True)
+loan_df['LoanAmount'].fillna(loan_df['LoanAmount'].median(), inplace=True)
+loan_df['Loan_Amount_Term'].fillna(loan_df['Loan_Amount_Term'].mode()[0], inplace=True)
+
+# Display dataset summary
+print("Loan Data Summary:\n")
+print(loan_df.describe())
+
+# Visualizations
+
+# Loan status count
+plt.figure(figsize=(6, 4))
+sns.countplot(data=loan_df, x='Loan_Status', palette='pastel')
+plt.title("Loan Approval Status")
+plt.xlabel("Loan Status")
+plt.ylabel("Count")
+plt.tight_layout()
+plt.show()
+
+# Loan status by property area
+plt.figure(figsize=(7, 5))
+sns.countplot(data=loan_df, x='Property_Area', hue='Loan_Status', palette='Set2')
+plt.title("Loan Status by Property Area")
+plt.xlabel("Property Area")
+plt.ylabel("Count")
+plt.legend(title='Loan Approved')
+plt.grid(axis='y', linestyle='--', alpha=0.3)
+plt.tight_layout()
+plt.show()
+
+# Applicant Income Distribution
+plt.figure(figsize=(7, 5))
+sns.histplot(loan_df['ApplicantIncome'], bins=25, kde=True, color='teal')
+plt.title("Distribution of Applicant Income")
+plt.xlabel("Applicant Income")
+plt.ylabel("Frequency")
+plt.grid(axis='y', linestyle='--', linewidth=0.5, alpha=0.7)
+plt.tight_layout()
+plt.show()
+
+# Credit history vs loan approval
+loan_df['Loan_Approved'] = loan_df['Loan_Status'].map({'Y': 1, 'N': 0})
+plt.figure(figsize=(6, 4))
+sns.barplot(data=loan_df, x='Credit_History', y='Loan_Approved', palette='muted')
+plt.title("Credit History vs Loan Approval")
+plt.xlabel("Credit History")
+plt.ylabel("Approval Probability")
+plt.grid(axis='y', linestyle='--', alpha=0.5)
+plt.tight_layout()
+plt.show()
+
+# Correlation heatmap
+plt.figure(figsize=(8, 6))
+cols = ['ApplicantIncome','CoapplicantIncome','LoanAmount','Loan_Amount_Term','Credit_History']
+corr_matrix = loan_df[cols].corr()
+sns.heatmap(corr_matrix, annot=True, cmap='YlGnBu', fmt=".2f", linewidths=0.5)
+plt.title("Correlation Matrix")
+plt.tight_layout()
+plt.show()