ML Project – Insurance Claim Approval using XGBoost Algorithm

Program 1

Insurance Claim Approval Dataset

# Step 1: Import required libraries
import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import LabelEncoder
from xgboost import XGBClassifier
from sklearn.metrics import accuracy_score, confusion_matrix
import matplotlib.pyplot as plt

# Step 2: Load dataset
df = pd.read_csv("D://scikit_data/insurancedata/insurance_claim_approval.csv")
df.head()
df.isnull().sum()
df.shape

# Step 3: Encode categorical features
label_encoders = {}
for col in ["Gender", "Smoking", "PolicyType", "PreExistingCondition"]:
    le = LabelEncoder()
    df[col] = le.fit_transform(df[col])
    label_encoders[col] = le  # Save encoders for later use
df.head()

# Step 4: Split features and target
X = df.drop("ClaimApproved", axis=1) # Input Data
y = df["ClaimApproved"] # output

# Step 5: Train-test split
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
len(X_test)

# Step 6: Initialize and train XGBoost model
model = XGBClassifier(use_label_encoder=False, eval_metric='logloss')
model.fit(X_train, y_train)

# Step 7: Predictions and Evaluation
y_pred = model.predict(X_test)
acc = accuracy_score(y_test, y_pred)
cm = confusion_matrix(y_test, y_pred)

print(" Model Accuracy:", round(acc * 100, 2), "%")
print("Confusion Matrix:\n", cm)

# Step 9: User Input Prediction
print("\n Enter details to predict claim approval:")

age = int(input("Age: "))
gender = input("Gender (Male/Female): ")
bmi = float(input("BMI: "))
smoking = input("Smoking (Yes/No): ")
policy = input("Policy Type (Basic/Premium/Gold): ")
amount = float(input("Claim Amount: "))
condition = input("Pre-existing Condition (None/Diabetes/Heart Disease/Asthma): ")
stay = int(input("Hospital Stay Days: "))

# Encode user inputs
gender = label_encoders["Gender"].transform([gender])[0]
smoking = label_encoders["Smoking"].transform([smoking])[0]
policy = label_encoders["PolicyType"].transform([policy])[0]
condition = label_encoders["PreExistingCondition"].transform([condition])[0]

input_data = pd.DataFrame([{
    "Age": age,
    "Gender": gender,
    "BMI": bmi,
    "Smoking": smoking,
    "PolicyType": policy,
    "ClaimAmount": amount,
    "PreExistingCondition": condition,
    "HospitalStayDays": stay
}])

# Predict
prediction = model.predict(input_data)[0]
print("\n Claim Status:", " Approved" if prediction == 1 else " Not Approved")

# Step 8: Feature Importance Plot
plt.figure(figsize=(10, 6))
plt.barh(X.columns, model.feature_importances_, color='skyblue')
plt.title(" Feature Importance - Claim Approval")
plt.xlabel("Importance Score")
plt.tight_layout()
plt.show()