About the project
Most people find out they overspent only when the month is over, because expenses are spread across UPI apps, cards, cash and paper bills, and typing every entry into a diary is tiring. This AI personal finance advisor and expense tracker Android project (PaisaPlan) removes most of the typing and turns the numbers into clear advice.
An expense can be added in three quick ways: scan the printed bill with the phone camera, share a bank / UPI SMS to the app, or type a short note. The app reads the shop, amount, date, bill number and payment mode, selects the category automatically and checks it against the user's budgets. At month end, Gemini AI studies the spending and replies with a financial health score, what it noticed and practical saving tips with the rupees each tip can save.
How it works
- Scan a bill: CameraX takes the photo and Google ML Kit reads the text on the phone itself
- Simple rules find the shop name, grand total, date, bill number and payment mode - every field stays editable
- Share any bank or UPI SMS to PaisaPlan and it opens a filled expense / income form
- Admin keywords pick the category automatically; Gemini is asked only when no keyword matches
- Monthly and category budgets raise an alert at the chosen % and again when the limit is crossed
- Gemini AI returns a health score (0-100), insights, saving tips and suggested budgets for next month; offline rules give the same advice without internet
- Family members spend from a shared wallet with monthly limits and can request money from the owner
Users
- Main user - records income and expenses, sets budgets, gets AI advice, tracks saving goals and owns the family wallet
- Family member - joins with an invite code, spends from the family wallet, scans bills and requests money
- Admin - manages users, categories and keywords, Gemini settings, tips and notices, and prints monthly reports
Highlights
- 3 roles, 19 Android screens, 29 JSON APIs, 14 admin panel pages, 12 MySQL tables
- Bill OCR runs on the device (bill photo is compressed and saved for the admin to verify)
- Duplicate bill check by bill number and amount
- AI works with a free Gemini API key; without a key the app still works with keyword categories and offline advice
- Charts (donut + 6-month bars) and a monthly PDF statement made on the phone
- Validations in the app and again in PHP, bcrypt passwords, token on every API call, prepared statements, CSRF check in admin forms
- Sample data with demo logins (main user, two family members, admin); sample dates follow the import day so the current month always has data
What you get
- Full Android Studio project (Java) and PHP admin panel + APIs source code
- MySQL database (.sql) with sample data
- Project report in Word (60+ pages) with UML, DFD, ER diagrams and test cases
- 2-page project abstract (Word + PDF)
- Project explanation with 30 viva questions
- Step-by-step setup guide (PDF), README and screenshots
- Setup help if you get stuck
Who can use this project
Suitable as a final year Android project with source code for BE / B.Tech (Computer, IT, ENTC), MCA, BCA and Diploma students, and as an M.Tech mini project. It covers useful topics for the viva: OCR, use of a Gemini AI model through an API, budget alert logic, a client-server Android app with a PHP / MySQL back end, and a bill scanner OCR Android project module you can explain step by step.
Frequently asked questions
Does the bill scan work without internet? Yes. ML Kit text recognition runs on the phone; internet is needed only to save the entry on your server and for Gemini advice.
Is a paid Gemini key needed? No. A free key from Google AI Studio is enough, and without any key the app gives offline rule-based advice.
Does the app read all my SMS? No. It reads only the message you share to it or paste, so no SMS permission is asked.
Can I run it on a real phone? Yes. Keep the phone and laptop on the same Wi-Fi and type the laptop IPv4 address on the login screen.
