risklytix

Risklytix, an advanced risk assessment platform, that harnesses the power of machine learning real-time stock data to assist users in precisely evaluating and mitigating risks.

Comment

🚀 Inspiration

Risk assessment is more important than ever in today’s digital world. We wanted to build something that helps users stay in control without drowning in complex data. Our goal? A smart, intuitive tool that makes risk assessment easy.

🔍 What it does

Risklytix calculates the risk with stocks on the NYSE, by using an LongShortTermMemory to detect trends with market data and company financials in collaboration with NLP sentiments. Risklytix displays a risk factor which is computed in real-time, so users can make the right decisions.

🛠️ How we built it

We combined the power of Python, machine learning, and cloud-based tools to bring Risklytix to life. The backend is built with Flask and Python, while the frontend is powered by React for a smooth user experience. First, we get the past 6-years company data and ticker information from the API and store it into an AWS database. Then, we then compute the necessary risk formulas for certain quarters/days in the past 6 years, and add that to the database as well. Then, we fed all this data into the LSTM,(trained the data first ) which then calculated the current risk given sentiment, past history, and time sequence trends. We then brought this data to the front-end where users can quickly see the risk assessment score of a ticker symbol.

⚠️ Challenges we ran into

  • Dealing with messy and incomplete datasets
  • Fine-tuning our LSTM, and trying to maximize accuracy
  • Connecting the backend to the frontend for user interaction
  • Small group size
  • Scalability and performance of the model- the model was performing slow.

🎉 Accomplishments that we're proud of

  • Achieving financial stability in risk assessment patterns. 🔥
  • Successfully building an LSTM that actually works!
  • Creating a user-friendly dashboard that makes complex insights easy to understand. 📊
  • Overcoming major integration challenges and ensuring seamless data processing. ✅

📚 What we learned

  • Full-stack—Being able to create a full-stack application that has a whole LSTM in the background!
  • We learned how to make an LSTM in coordination with NLP sentiments
  • Use AWS, and specifically manage Amazon DynamoDB
  • Updated our financial literacy

🔮 What's next for Risklytix

  • Making our LSTM even smarter by feeding it more data. 🤯
  • Offering API integrations so businesses can seamlessly embed Risklytix into their workflows. 🔗
  • Hosting the site so the everyone has access to it!
Share this project:

Updates