Dummy Stocks

Demo

Inspiration

As stock trading gains popularity, more Americans are diving in—often influenced by trends or advice from influencers. Unfortunately, many end up losing money by following the hype rather than truly understanding the companies they invest in. A more reliable way to evaluate a company’s growth would be through financial articles. But these are often filled with complex jargon that can intimidate new traders trying to make sense of the market.

What it does

Our application allows users to input a company they’re interested in, and it pulls up relevant news articles. Using a large language model, Dummy Stocks simplifies these articles, making complicated financial language filled articles easy for beginners to understand. This gives new traders a clear, up-to-date view of a company's situation, helping them make smarter investment choices. On top of this, Dummy Stocks offers a visual comparison tool: a scatter plot showing the revenue of the chosen company alongside others in the market. Users can hover over each point to see company names and instantly understand how one company compares to others.

How we built it

The application uses a Flask backend and a React Vite frontend. When a user enters a stock name in the search bar, the app searches for articles related to the stock using the NewsAPI. If an article contains the user input, it is displayed on the screen, and the content is then simplified by passing it through an OpenAI model before showing it to the user. The scatter plot is generated using ChartJS, with financial data pulled from the FinnhubAPI. The frontend is deployed on Vercel, while the backend is hosted on Render.

Challenges we ran into

A major challenge on the article side was that many news providers, like Yahoo, had web-scraping blockers that made it difficult to access the articles we needed. Finding free APIs that met our project scope and allowed a reasonable number of calls was also time-consuming. Additionally, deploying the app was challenging, as several deployment platforms didn’t support deploying a Flask backend, which required us to explore alternatives.

Accomplishments that we're proud of

Accomplishments we are proud of is implementing multiple different API and implementing the OpenAI LLM into the solution for our problem in simplifying the text. Another thing we are proud of is connecting to the many APIs that were used in this program and learning about stocks, as we were amateurs in investing. Additionally, the clean front-end is also something to be proud of as it took a lot of effort.

What we learned

For starters this application forced us to work with many different technologies and managing and learning on the fly to debug. We learned many frameworks such as React, Flask, OpenAI, and APIs.

What's next for Dummy Stocks

In the future, we plan to implement more data visualization options to help users understand how certain news and data affect a company. For example, we’re looking to add line charts to show trends over time and pie charts to visualize a company’s market cap percentage. By incorporating more simple, informative graphs, we aim to provide users with even more insights, enabling them to make smarter financial decisions in the stock market.