Finance Sis

The risk assessment score and chatbot

Inspiration

Financial scams disproportionately target women and college students, especially through emails, job offers, investment “opportunities,” and payment requests, especially for tuition and housing, that feel urgent. Many existing scam detectors feel intimidating or overly technical, which can discourage people from trusting or using them. We wanted to create a tool that feels supportive, accessible, and empowering to people who are learning about managing their finances. Finance Sis was inspired by the idea that financial literacy and safety should feel like advice from a trusted friend, not a warning label.

What it does

Finance Sis is a financial scam detector that analyzes messages such as emails or texts and assigns a risk score based on scam indicators, including the email address and message contents. It uses natural language processing to identify suspicious keywords, phrasing patterns, and behavioral red flags commonly found in financial scams. After analysis, the chatbot explains why something may be risky in clear, non-technical language, helping users make informed decisions rather than just labeling something as “safe” or “unsafe.” It also advises users on general financial health.

How we built it

We built the frontend using React, connecting it to our Python backend using Flask. The incoming data from users was processed and scored, adding to the score when key phrases were identified or when the email address was deemed risky.

Challenges we ran into

We had trouble using our API to power our chatbot. We also came across issues connecting our frontend and backend.

Accomplishments that we're proud of

We are proud of the look of the website, and the connections were able to make between the backend and frontend. We are also proud of the concept and how simple user input can yield meaningful results.

What we learned

We learned about connecting various components of a website. We also learned how to guide a project starting from ideation to a working end product. Lastly, we learned about natural language processing and the process of making a chatbot.

What's next for Finance Sis

We want to process more scam data from outside sources to better analyze messages. We also want to use the contents of the message to identify the scam type. Furthermore, we want to offer safer financial alternatives to the scammer's proposed opportunity. We hope to later incorporate the identification of financial risk based on cost amounts compared to the user's financial situation rather than solely the scam risk.