Skip to content

πŸš€ This project analyzes 2.5+ lakh UPI transactions worth β‚Ή32.79 Cr to uncover trends, fraud risks, and user behavior. Built with Tableau dashboards & reports, it highlights peak usage, category insights, fraud hotspots, and provides actionable recommendations to improve security, trust, and performance.

Notifications You must be signed in to change notification settings

gulnaaz-data-analyst/UPI-Transaction-Analysis-and-Fraud-Detection

Repository files navigation

πŸ“Š UPI Transactions Analysis And Fraud Detection

πŸ“Œ Project Overview

UPI (Unified Payments Interface) is one of the fastest-growing digital payment systems in India. While adoption is rising, so are fraud risks.
This project analyzes 2.5+ lakh UPI transactions worth β‚Ή32.79 Cr to uncover trends, fraud patterns, and customer behavior, while providing recommendations to strengthen security, performance, and user trust.

The repository includes:

  • πŸ“Š UPI-Transactions-and-Fraud-Analysis.twbx β†’ Tableau Dashboard for interactive insights
  • πŸ“‘ UPI-Transactions-and-Fraud-Analysis.pptx β†’ Presentation summarizing findings & recommendations

🎯 Problem Statement

The surge in UPI usage brings challenges:

  • Increasing fraud incidents in certain networks, devices, and banks
  • Lack of visibility into peak usage, demographics, and bank-wise performance
  • Difficulty for businesses to improve services, reduce fraud, and build trust

πŸ“Œ Objectives

  • Analyze UPI transaction trends across banks, devices, and merchant categories
  • Detect & monitor fraud patterns, especially high-value and suspicious activity
  • Understand user behavior across age groups, locations, and time of usage
  • Deliver data-driven insights for security, optimization, and better decision-making

πŸ“‚ Data Overview

The dataset combines transactions, demographics, and fraud details:

  • Transactions β†’ ID, Amount, Date-Time, Status, Weekend/Weekday
  • Demographics β†’ Age Group, Bank Name, Sender & Receiver
  • Fraud Details β†’ Network Type, Device Type, Fraud Flag

πŸ” Key Insights

  • Transaction Patterns:

    • Evenings (6–9 PM) = peak usage
    • Weekdays β†’ higher transaction volume; Weekends β†’ higher average value
    • Top categories: Shopping & Grocery
    • Success rate: 95%, failures peak during rush hours
  • Fraud Risks:

    • 480 suspicious transactions flagged
    • Grocery, Food, Shopping = most targeted categories
    • Android users face more fraud risk than iOS
    • 4G/5G networks show higher fraud than WiFi/3G
    • Young adults (18–30 yrs) most affected
    • SBI Bank reported higher fraud cases

βœ… Recommendations

  • Strengthen authentication (biometrics / OTP) for high-value payments
  • Partner with telecom providers to monitor fraud-prone 4G/5G activity
  • Run awareness campaigns for young adults (18–30 yrs)
  • Provide real-time fraud alerts & one-click reporting in UPI apps
  • Upgrade fraud detection in high-risk banks & merchants
  • Optimize server loads during evening peaks to reduce failures
  • Launch cashback/reward programs for off-peak usage

πŸ’‘ Business Impact

  • πŸ”’ Stronger fraud detection & prevention β†’ lower financial losses
  • 🀝 Higher customer trust & adoption β†’ sustained UPI growth
  • ⚑ Fewer failures β†’ better performance & reduced support costs
  • πŸ“œ Improved regulatory compliance
  • 🎯 Personalized experiences β†’ higher engagement & retention
  • πŸ“ˆ Increased transaction volumes & revenue growth
  • πŸ† Competitive edge for secure & reliable UPI providers

πŸ“Š UPI Transaction Analysis Dashboard

Here’s a snapshot of my dashboard:

Dashboard Screenshot

πŸ“Š UPI Transaction Analysis & Fraud Detection

Dashboard Preview

UPI Dashboard

About

πŸš€ This project analyzes 2.5+ lakh UPI transactions worth β‚Ή32.79 Cr to uncover trends, fraud risks, and user behavior. Built with Tableau dashboards & reports, it highlights peak usage, category insights, fraud hotspots, and provides actionable recommendations to improve security, trust, and performance.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published