import pandas as pd
import numpy as np
from sklearn.ensemble import IsolationForest
from sklearn.preprocessing import StandardScaler
import matplotlib.pyplot as plt
 
# Sample stock market transactions data
# Let's assume the dataset has features like transaction ID, stock symbol, price, volume, and time
data = {
    'TransactionID': [1, 2, 3, 4, 5, 6, 7, 8, 9, 10],
    'Price': [100, 105, 200, 150, 110, 107, 130, 155, 5000, 108],
    'Volume': [1000, 1100, 1200, 1000, 1300, 1100, 1400, 1500, 12000, 1050],
    'Time': [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]  # Simple time variable for illustration
}
 
# Convert to a DataFrame
df = pd.DataFrame(data)
 
# Feature selection: Using Price and Volume to detect fraud
features = df[['Price', 'Volume']]
 
# Standardizing the data (important for anomaly detection)
scaler = StandardScaler()
scaled_features = scaler.fit_transform(features)
 
# Using Isolation Forest for anomaly detection
model = IsolationForest(contamination=0.2)  # 20% of the transactions could be fraudulent
model.fit(scaled_features)
 
# Predict anomalies (1 is normal, -1 is anomalous)
df['Anomaly'] = model.predict(scaled_features)
 
# Displaying results
print("Transaction Data with Anomaly Detection:")
print(df)
 
# Visualizing the result
plt.figure(figsize=(10, 6))
plt.scatter(df['Price'], df['Volume'], c=df['Anomaly'], cmap='coolwarm', s=100, edgecolors='black')
plt.title('Stock Market Transactions with Fraud Detection')
plt.xlabel('Price')
plt.ylabel('Volume')
plt.colorbar(label='Anomaly')
plt.show()

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