🧾 Fraud Detection System

Overview :

This project builds a machine learning model to detect fraudulent transactions based on behavioral and transactional features. It supports financial institutions in identifying suspicious activity and preventing fraud in real time.

---

Dataset Description :

The dataset includes anonymized transaction records with engineered features for fraud detection.

Key columns:

• Time: Time elapsed since the first transaction
• Amount: Transaction amount
• V1 to V28: PCA-transformed features capturing behavioral patterns
• Class: Target label (1 = Fraud, 0 = Legitimate)

---

Workflow Summary :

1. Data Loading

df = pd.read_csv('/kaggle/input/fraud-detection-dataset/fraud.csv')

2. Preprocessing

df = df.dropna()
X = df.drop('Class', axis=1)
y = df['Class']

3. Train-Test Split

X_train, X_test, y_train, y_test = train_test_split(X, y, stratify=y, test_size=0.2, random_state=42)

4. Model Training

model = RandomForestClassifier(n_estimators=100, random_state=42)
model.fit(X_train, y_train)

5. Evaluation

y_pred = model.predict(X_test)
print(accuracy_score(y_test, y_pred))
print(classification_report(y_test, y_pred))
print(confusion_matrix(y_test, y_pred))

6. Prediction on New Transaction

input_df = pd.DataFrame(np.zeros((1, len(X_train.columns))), columns=X_train.columns)
model.predict(input_df)

---

Performance Metrics :

• Accuracy: ~99% on test data
• Precision: High precision for fraud class
• Top predictors: PCA components V14, V17, V10, and transaction amount

---

Dependencies :

numpy
pandas
scikit-learn

---

Author: Sushree Bandita Das

---