AInvest

• 

  Home

• 

  console log of training data and classification

• 

  example ticker lookup with articles and news score

• 

  sector scores page

Inspiration

We were inspired by the complicated field of finance and wanted to prevent its complications from being an obstacle for potential investors. One of our founders also lost 40% of his portfolio in three months, so he was motivated by rage to work on this project. He'e broke anyway.

What it does

AInvest uses a combination of APIs to retrieve news articles and analyze them using a custom sentiment analysis model. It then uses this to recommend sectors within markets that may have potential for growth and warn of unexpected market changes.

How we built it

We used NodeJS and React along with Axios as our direct technologies. We interface with several APIs, including Fidelity Investments' charts and stock information and Fidelity News by stock ticker in order to provide relevant information. We also conducted research into financial markets in order to provide relevant analysis of data and train our detection model.

Challenges we ran into

We ran into significant challenges with standardization across APIs. Timescales, formats, and data are often found in numerous forms which required us to write parsing functions in order to standardize data for use. We also overcame significant challenges in terms of training and using the AI to detect positivity in the stock market as financial terms are not in common use and thus not easily detectable by pre-trained sentiment analysis systems.

Accomplishments that we're proud of

We are proud to have completed and fully integrated the sentiment analysis system into a working web app that is easily accessible to users. We are also very proud of the degree of research conducted for this project, since we learned a lot more about financial risk analysis techniques while doing our project.

What we learned

We learned the importance of teamwork (yes, it's cheesy) as we focused on integrating many different pieces of the project together - finance and technology, my code and your code, our project and the APIs, etc.

What's next for AInvest

We want to continue working on AInvest since we believe its approach is underutilized in the financial sector and AInvest can provide useful information for everyday investors. We plan to add more tools that ease the learning curve and allow more people to invest knowledgable. Namely, we plan to add a text summarization AI that condenses and rewords technical articles to remove the obstacle of formal jargon for investments. Additionally, we want to add several other financial risk analysis tools, such as the efficient frontier and the Modern Portfolio Theory, the VaR (value-at-risk), and historical means and standard deviations. Lastly, we want to make sure the UI looks cleaner, by adding cards and soft-fading transitions between elements on the website.

Built With

• axios
• fidelity
• javascript
• node.js
• react