As our world shifts to mobile payments, it's becoming harder and harder to budget our finances and differentiate between luxuries and necessities.
What it does
A simple financial wellness application that allows you to digitize your receipts and leverage state-of-the-art Machine Learning and Natural Language Processing to categorize your budget into unique categories.
How we built it
- React and Material-UI frameworks for development of authentication, transactions, and budget user interfaces
- Firebase API for user authentication and cloud integration
- Google Cloud Firestore for receipt item inventory, hosting images, and hosting client applications.
- Google Cloud Vision OCR for conversion of image text into machine-encoded text
- Google Cloud NLP Content Classifier for classification of receipt items into a taxonomy of unique shopping categories
- Google Cloud App Engine to host server which interacts with the client application hosted on Firebase
Challenges we ran into
- Implementation of the OCR to register the correct information
- Content categorization
- Synchronizing API calls with promises and daisy-chaining
Accomplishments we're proud of
- A simple and friendly user interface
- Perfect carry-out and usage of multiple API's
- Application of Machine Learning to assign receipt items an appropriate category
What we learned
We learned a lot of technical skills such as api implementation, web server hosting, and front-end vs. back-end design.
What's next for SimpliFi
We believe SimpliFi has much more potential for additional features.
Location and Fraud
- Use Google Maps API to keep a record of where all purchases were made
- Purchase made in another country will alert user of possible fraud
User Spending Habits
- Use machine learning to determine the best budgeting plan based on the user's spending habits