CreditLens

Risk Analysis
Top 2 Investible Companies Description
5 Years Growth Predictions
Credit Score Based Investment Allocation Strategy

Inspiration

The inspiration behind CreditLens came from the limitations of traditional investment analysis methods that rely heavily on a single metric such as credit score. In real-world financial decision-making, investors consider multiple factors including financial stability, revenue performance, and risk exposure. This project was inspired by the goal of building a more comprehensive, data-driven system that combines machine learning with financial fundamentals to support smarter and more balanced investment decisions.

What it does

CreditLens analyzes companies using multiple financial parameters such as credit score, financial health, revenue, and risk indicators. It generates an overall investment score that is used to rank companies, allocate investment percentages, compare top-performing firms, predict risk levels, and forecast company growth over the next five years. The system provides clear visualizations and insights to support both short-term and long-term investment strategies.

How we built it

We built CreditLens using a Linear Regression model trained on historical financial data. The model takes multiple financial inputs and predicts key outcomes such as credit score trends, risk levels, and growth potential. These predictions are combined into a weighted investment score that drives allocation and ranking decisions. The trained model is stored as linear_regression_new_model.pkl, and external market indicators are integrated using the Kalshi API. Data visualizations such as pie charts, bar charts, radar charts, and growth trend lines are used to present the results effectively.

Challenges we ran into

One of the biggest challenges was integrating multiple financial parameters into a single, balanced investment score without overemphasizing any one factor. Risk prediction was particularly challenging due to the influence of external market conditions. Ensuring that long-term growth forecasts aligned with realistic financial behavior also required careful model tuning and validation.

Accomplishments that we're proud of

We are proud of building a complete end-to-end investment analysis system that goes beyond credit scores and incorporates risk-aware decision-making. Successfully integrating machine learning predictions with real market indicators and presenting them through clear, intuitive visualizations was a major achievement. The project demonstrates a strong balance between model interpretability and practical financial insight.

What we learned

Through this project, we learned how to apply machine learning models to financial data in a responsible and interpretable way. We gained experience in feature engineering, risk modeling, API integration, and financial data visualization. We also learned the importance of validating assumptions and designing models that align with real-world financial reasoning.

What's next for CreditLens

Next, we plan to enhance CreditLens by incorporating additional machine learning models for improved prediction accuracy, such as ensemble methods. We also aim to include real-time data pipelines, advanced risk modeling techniques, and scenario-based stress testing. Expanding the platform into an interactive dashboard and supporting broader asset classes are also part of our future roadmap.