Inspiration
The inspiration for MarketLens came from observing how overwhelming and fragmented stock market information can be, especially for students and non-experts. While large amounts of public financial data exist, interpreting earnings reports, price movements, and market news together is challenging. This project was motivated by the idea that AI could help synthesise recent public data into clear, explainable insights about company performance, risk, and long-term viability, supporting more informed and responsible analysis.
What it does
MarketLens analyses public, recent data from top global companies to assess their market performance, investment risk, and likelihood of long-term success. It combines quantitative indicators such as stock trends, volatility, and financial health with qualitative signals like news sentiment. The system outputs risk scores, success probability estimates, and natural-language explanations, acting as a decision-support tool rather than a source of investment advice.
How we built it
The project was built around AI, the core reasoning engine. Public datasets such as historical stock prices, financial statements, and recent news articles were first processed to extract meaningful features. These included growth metrics, volatility measures, and sentiment scores.
A structured scoring framework was then designed, where AI analysed relationships between features to generate interpretable insights. Finally, the results were presented in a clear, user-friendly format, highlighting both insights and limitations.
Challenges we ran into
One major challenge was avoiding overconfidence in predictions. Financial markets are influenced by unpredictable external events that cannot be fully captured by data. Another challenge was ensuring that the model did not confuse correlation with causation, especially when analysing news sentiment. Balancing analytical depth with explainability was also difficult, as overly complex models can be hard for users to trust or understand.
Accomplishments that we're proud of
We are proud of building a system that integrates numerical data with AI-driven qualitative reasoning in a transparent way. Instead of producing black-box predictions, MarketLens explains why certain risks or opportunities are highlighted. We also successfully framed the project around ethical AI use, clearly communicating uncertainty and avoiding misleading claims.
What we learned
This project deepened our understanding of financial analysis, machine learning concepts, and responsible AI design. We learned that AI outputs should be interpreted probabilistically rather than as certainties, and that data quality and recency play a critical role in meaningful analysis. Most importantly, we learned that explainability is just as important as accuracy when building AI systems for real-world decision support.
What's next for MarketLens
Future improvements include expanding scenario analysis under different economic conditions, adding longer-term backtesting to evaluate robustness, and enhancing visualisation features for clearer comparisons between companies. With further development, MarketLens could become a powerful educational tool for understanding how data-driven reasoning can be applied responsibly in financial contexts.
Log in or sign up for Devpost to join the conversation.