Credit Card Fraud Detection System Final Year Project with Source Code

Dashboard Features in this Final Year Project

• Professional sidebar layout
• Transaction summary cards
• Total transaction count
• Checked transaction count
• Safe transaction count
• Fraud transaction count
• Suspicious transaction count
• Recent checked transactions
• Latest fraud status display
• Notification history
• Quick action links
• Upload transaction shortcut
• Add manual transaction shortcut
• Report download shortcut

Transaction Management Features in this Final Year Project

• Manual transaction entry
• CSV transaction upload
• Row-level CSV validation
• Invalid row error report
• Pending transaction status
• Fraud checked transaction status
• Single transaction fraud check
• Bulk transaction fraud check
• Check all pending transactions
• Transaction details page
• Transaction delete with confirmation
• Batch delete support
• Per-user transaction privacy
• User can only view own transactions
• Export and clear data controls

Fraud Detection Logic in this Final Year Project

This project uses a rule-based fraud scoring system. It does not require real banking datasets or external banking APIs.

Risk Factors

Final Classification Rules

Example

Amount high       → +30
Late night        → +20
Crypto merchant   → +25
-------------------------
Total Risk Score  → 75
Final Status      → Fraud

Fraud Result Details Features in this Final Year Project

For every checked transaction, the user can view:

• Full transaction information
• Transaction amount
• Date and time
• Merchant name
• Transaction type
• Location
• Card profile
• Final fraud status
• Risk score
• Reason tags
• Check date and time
• Fraud explanation
• Human-readable risk reason

Example reason labels:

• Unusual transaction amount
• Transaction at odd hours
• High-risk merchant detected
• Daily limit exceeded
• Rapid transaction frequency detected
• Unusual location pattern

Internal Risk Analysis Modules in this Final Year Project

1. Amount Analysis

• Compares transaction amount with usual spending range
• Checks daily card limit if card profile is available
• Flags unusually high transactions
• Adds higher risk when amount crosses defined limit

2. Time-Based Pattern Check

• Detects late-night transactions
• Flags transactions between 12 AM and 5 AM
• Adds risk when transaction time looks unusual

3. Merchant Risk Pattern

• Detects high-risk merchant keywords
• Flags crypto platforms
• Flags gift card or voucher merchants
• Flags unknown or rarely used merchants

4. Location Deviation

• Checks unusual transaction location if available
• Adds risk for unknown or sudden location change
• Useful for fraud simulation logic

5. Frequency and Velocity Check

• Detects multiple transactions in short time
• Flags repeated transactions within minutes
• Adds risk for rapid retries or transaction bursts

Card Profile Features in this Final Year Project

• Add card nickname
• Add daily limit
• Add usual spending range
• Assign card profile to transaction
• Use card profile for risk calculation
• Detect daily limit exceeded
• Detect spending beyond normal range
• Improve transaction risk explanation
• Make project more complete without external banking integration

Report Features in this Final Year Project

Users can download reports in:

• PDF format
• CSV format

Report types:

• All transactions report
• Fraudulent transactions report
• Suspicious transactions report
• Custom date range report
• User transaction summary report

Report includes:

• User details
• Summary counts
• Transaction list
• Fraud status
• Risk score
• Reason tags
• Generated date and time

Notification Features in this Final Year Project

• In-app notification system
• Bulk check completed notification
• Fraud transaction alert
• Suspicious transaction alert
• Daily fraud summary message
• Dashboard notification display
• Notification history

Example notifications:

• 3 transactions marked Fraud today
• Bulk check completed
• CSV upload completed with 2 invalid rows
• Transaction report downloaded successfully

Help and Guidelines Features in this Final Year Project

• CSV format sample
• Upload instructions
• Fraud check steps
• Report download steps
• Common upload errors
• Common validation errors
• Transaction field explanation
• Safe/Suspicious/Fraud meaning
• User tips for better transaction tracking

Sample CSV file:

sample_data/sample_transactions.csv

Data Privacy Features in this Final Year Project

• Per-user data visibility
• User can view only own transactions
• User can delete own transactions
• User can delete uploaded batch records
• User can export data
• User can clear all personal data
• No external banking API required
• Fully local project execution
• Safe for academic demonstration

• User registration in this final year project
• User login
• User logout
• Forgot password using security question
• Forgot password using system-generated OTP
• Profile management
• Update full name
• Update email
• Update mobile number
• Change password with old password confirmation
• Dashboard with sidebar layout
• View total transactions uploaded/added
• View total checked transactions
• View fraud flag count
• View safe transaction count
• View recent activity
• Add transaction manually
• Enter transaction amount
• Enter transaction date
• Enter transaction time
• Enter merchant/shop name
• Select transaction type: online/offline
• Add location
• Add card type
• Add notes
• Validate transaction amount
• Validate transaction date and time
• Save transaction as pending check
• Upload transactions using CSV
• Validate CSV file format
• Validate required columns
• Skip invalid rows
• Generate row-level error report
• Store valid uploaded transactions
• Check single transaction
• Check multiple selected transactions
• Check all pending transactions
• View Safe/Suspicious/Fraud status
• View risk score
• View reason labels
• View full fraud result details
• Search transactions by merchant name
• Search transactions by transaction ID
• Filter transactions by status
• Filter transactions by date range
• Filter transactions by amount range
• Sort transactions by latest first
• Sort transactions by highest amount
• Sort transactions by risk score
• Download reports in PDF
• Download reports in CSV
• Download all transaction reports
• Download suspicious/fraud-only reports
• Download custom date-range reports
• Add optional card profiles
• Assign transaction to card profile
• View in-app notifications
• Delete own transactions
• Delete uploaded batch records
• Export user data
• Clear own data
• Access help and guidelines page

Machine Learning Note in this Final Year Project

This version does not require a real machine learning model.

The project intentionally uses a rule-based fraud detection logic because:

• Real fraud ML requires large private banking datasets
• Banking datasets are highly sensitive
• Rule-based fraud screening is still used in real-world systems
• It is easier to explain in viva
• It works fully offline
• It is safe for academic projects

ML-ready explanation:
The risk score can later be used as a feature set for Logistic Regression, Random Forest, or other machine learning models.

Other Features in this Final Year Project

• Complete Credit Card Fraud Detection source code
• Suitable for final year project, major project, and minor project
• Project report content can be prepared from included modules
• Flask web application
• SQLite database
• User authentication
• Forgot password module
• Security question verification
• OTP generated inside system
• Dashboard with sidebar layout
• Manual transaction entry
• CSV transaction upload
• Row-level CSV validation
• Fraud risk scoring
• Safe/Suspicious/Fraud classification
• Result details page
• Search/filter/sort transactions
• PDF report generation
• CSV report generation
• Optional card profiles
• In-app notifications
• Data privacy controls
• Help page
• Sample transaction CSV
• Demo seed data
• Useful for viva, source code review, project report writing, and project demonstration

• Open terminal in the project folder:

  cd "credit-card-fraud-detection-python-web"

• Create virtual environment:

  python -m venv venv

• Activate virtual environment on Windows:

  venv\Scripts\activate

• Activate virtual environment on Linux/macOS:

  source venv/bin/activate

• Install dependencies:

  pip install -r requirements.txt

• Run the application:

  python run.py

• Open application:

  http://127.0.0.1:5000/

• Optional: seed dummy data:

  python scripts/seed.py

• Use sample CSV for testing:

  sample_data/sample_transactions.csv

Demo User Account after seeding

• Username: demo
• Email: [email protected]
• Password: Demo123

Seed data creates:

• 1 demo user
• 8 transactions
• 2 card profiles