Visualisations, Market Research and generated data

neo_dolfin/ai/generated_data/loan.py

@@ -0,0 +1,32 @@
+import pandas as pd
+import numpy as np
+import random
+from datetime import datetime, timedelta
+
+# Define variables
+start_date = datetime(2020, 1, 1)
+end_date = datetime(2021, 12, 31)
+account_number = 'ACC123456789'  # Sample account number
+
+# Generate mock data for the Transaction table for a single user
+transaction_data = pd.DataFrame({
+    'Transaction ID': ['TRX' + ''.join(random.choices('0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ', k=8)) for _ in range(1000)],
+    'Account number': [account_number for _ in range(1000)],
+    'Transaction date': [start_date + timedelta(days=random.randint(1, 730)) for _ in range(1000)],
+    'Transaction time': [datetime.strptime(f"{random.randint(0,23):02d}:{random.randint(0,59):02d}", "%H:%M").time() for _ in range(1000)],
+    'Transaction amount': [random.uniform(-1000, 1000) for _ in range(1000)],
+    'Credit or debit indicator': [random.choice(['Credit', 'Debit']) for _ in range(1000)],
+    'Statement description': ['Transaction ' + str(i) for i in range(1000)],
+    'High level transaction purpose': [random.choice(['Shopping', 'Dining', 'Utilities', 'Transportation']) for _ in range(1000)],
+    'Mid level transaction purpose': [random.choice(['Online Shopping', 'Restaurant', 'Electricity', 'Gas']) for _ in range(1000)],
+    'Low level transaction purpose': [random.choice(['Amazon', 'McDonalds', 'Electricity bill', 'Gas bill']) for _ in range(1000)],
+    'Transaction code': [random.randint(1000, 9999) for _ in range(1000)],
+    'Merchant ID': [random.randint(100000, 999999) for _ in range(1000)],
+    'Previous repayment date': [start_date - timedelta(days=random.randint(1, 30)) for _ in range(1000)]
+})
+
+# Update the product type to include random products
+transaction_data['Product type'] = [random.choice(['appliances', 'clothing', 'electronics', 'groceries', 'furniture']) for _ in range(1000)]
+
+# Save data to Excel sheet
+transaction_data.to_excel('loan.xlsx', index=False)

neo_dolfin/ai/generated_data/personal_loan_data.py

@@ -0,0 +1,54 @@
+import pandas as pd
+import numpy as np
+import random
+from datetime import datetime, timedelta
+
+# Define variables
+monthly_due_date = 5  # Assume the due date is the 5th of each month
+monthly_repayment_amount = 100  # Monthly repayment amount in currency
+transaction_limit = 1000  # Transaction/account limit in currency
+interest_value = 10  # Interest value for late payment in currency
+late_payment_value = 50  # Penalty for late payment in currency
+loan_amount = 5000  # Loan amount in currency
+
+# Generate personal loan dataset
+def generate_personal_loan_data(start_date, end_date):
+    dates = pd.date_range(start=start_date, end=end_date, freq='MS')
+    data = {'Date of Transaction': [], 'Description': [], 'Amount': [], 'Balance of Account': [], 'User ID': []}
+    balance = loan_amount
+    for date in dates:
+        # Check if it's the due date
+        if date.day == monthly_due_date:
+            # Assume a random chance of late payment
+            if random.random() < 0.5:  # 50% chance of late payment
+                description = 'Late Payment'
+                amount = -(monthly_repayment_amount + late_payment_value)
+                balance -= monthly_repayment_amount + late_payment_value
+            else:
+                description = 'Repayment'
+                amount = -monthly_repayment_amount
+                balance -= monthly_repayment_amount
+        else:
+            # Assume random transactions throughout the month
+            description = 'Charge' if random.random() < 0.5 else 'Transaction'
+            amount = random.uniform(-transaction_limit, transaction_limit)
+            balance += amount
+
+        data['Date of Transaction'].append(date)
+        data['Description'].append(description)
+        data['Amount'].append(amount)
+        data['Balance of Account'].append(balance)
+        data['User ID'].append(random.randint(10000, 99999))  # Generate random user ID
+
+    return pd.DataFrame(data)
+
+# Generate data for two years (24 months)
+start_date = datetime(2022, 1, 1)
+end_date = datetime(2023, 12, 31)
+personal_loan_data = generate_personal_loan_data(start_date, end_date)
+
+# Flag user for debt collection if amount is negative
+personal_loan_data['Debt Collection Flag'] = np.where(personal_loan_data['Amount'] < 0, 'Yes', 'No')
+
+# Save data to Excel
+personal_loan_data.to_excel('personal_loan_data.xlsx', index=False)