Inspiration
Financial markets react strongly to news, yet understanding the overall market mood from scattered headlines is difficult. This project was inspired by the idea of converting unstructured financial news into a clear, interpretable sentiment score that reflects market psychology in real time.
What it does
Market Pulse analyzes real-time financial news related to a stock or index and determines the overall market sentiment. It produces a sentiment label (positive, negative, or neutral), a numerical sentiment score, and the number of articles analyzed to give users a concise market overview.
How we built it
The backend was developed using FastAPI to fetch live financial news and process it through a deterministic sentiment-scoring engine. A rule-based model assigns sentiment scores in the range [-1 to 1] based on financial keywords. The frontend was built with Streamlit, enabling users to interactively query tickers and visualize sentiment results. The system was deployed using cloud-based services for accessibility.
Challenges we ran into
Handling noisy and ambiguous financial headlines was a major challenge, especially for market indices where sentiment is often subtle. Another challenge was ensuring stable and consistent sentiment scores while integrating intelligent analysis. Debugging deployment issues and frontend–backend communication also required careful attention.
Accomplishments that we're proud of
We successfully built a full-stack, real-time market sentiment analysis system with stable and explainable outputs. The project demonstrates a practical and reliable approach to sentiment analysis that works consistently with live data.
What we learned
We gained hands-on experience in full-stack development, API integration, cloud deployment, and debugging distributed systems. The project also taught us the importance of combining deterministic logic with intelligent systems to achieve reliable and interpretable results in real-world applications.
Log in or sign up for Devpost to join the conversation.