-
Search your favourite stock for real time analysis - it can be any stock which is listed on US stock market.
-
Latest financial News of your selected company with Sentiment Analysis
-
Overview of the company, competitors and profit growth
-
Get a Discounted Cash Flow Model of the company, understand its valuation and what it means for future stock price.
-
Key Financial Ratios - Liquidity, Debt etc. See a company's financial health based on a complex scoring system. Get insights on the same.
-
Play with interactive plotly visualizations.
-
See live stock price and a Benchmark method of Fast and Slow Moving Averages to get buy or sell signal for a stock.
-
LSTM based Machine Learning Model for future price prediction.
-
And if you still do not understand anything - our Chatbot fingenie is there to help.
Inspiration
We were inspired by the challenges faced by individuals without a finance background who often feel lost when trying to understand financial data and metrics. With Finance101, we set out to simplify the experience by creating a one-stop platform. Users can select a stock ticker to access tailored financial news, company health insights, stock analysis, valuations, and more about that ticker—all in one place. At the heart of the app is an AI-powered chatbot designed to answer questions, from the basics to advanced topics, ensuring that finance becomes accessible, engaging, and empowering for everyone.
What it does
This project offers an in-depth financial analysis suite with five specialized dashboards and an interactive finance chatbot based on live data from yfinance API. Central to the experience is the Financial Health Dashboard, which features a robust, custom-built scoring system that translates raw financial metrics into a clear health score. This scoring system evaluates liquidity, debt, efficiency, and profitability ratios, using a weighted, point-based approach to rate a company’s financial health. Each metric, from Current and Quick Ratios to Return on Equity (ROE), is assessed for its contribution to stability, with scores assigned based on performance thresholds, providing users with a transparent, data-driven financial health status.
The suite also includes the Overview Dashboard for key fundamentals, a Valuation Dashboard leveraging Discounted Cash Flow (DCF) analysis to gauge intrinsic share value, a Stock Predictions Dashboard with Moving Average and LSTM-based forecasts, and a Financial News Dashboard that captures real-time headlines and sentiment. Enhanced with Plotly, Streamlit, Tensorflow and Keras, the platform translates complex financial data into an intuitive, engaging experience. A finance chatbot is always available, ready to answer questions, making this tool a holistic companion for investors and financial enthusiasts seeking informed, actionable insights.
How we built it
We developed our financial analysis platform using a robust, multi-layered methodology to ensure data accuracy, predictive precision, and seamless user experience. We use the yFinance API to extract live financial data, processed with pandas for cleaning, normalization, and efficient handling. The entire system is powered by Streamlit for an interactive and dynamic web-based interface, enabling seamless navigation and user engagement.
Our Financial Health Dashboard features a custom scoring system, evaluating liquidity, debt, efficiency, and profitability through a weighted, point-based approach. Metrics like the Current Ratio and ROE are rigorously analyzed against benchmarks to deliver a transparent health assessment.
The Valuation Dashboard employs Discounted Cash Flow (DCF) analysis, forecasting Free Cash Flows and calculating Terminal Value, discounted using WACC to estimate intrinsic share value. Our Stock Predictions Dashboard leverages LSTM models built in Keras and TensorFlow, using historical data to forecast prices, complemented by Moving Average signals for buy/sell insights.
Plotly provides interactive visualizations. The Financial News Dashboard uses Stock News API with sentiment analysis to keep users informed, and our finance chatbot, powered by NLP, answers queries in real-time, offering financial guidance. This comprehensive, tech-driven approach makes our platform an indispensable tool for investors and analysts.
Challenges we ran into
We faced several significant challenges throughout the development process. First, building and training the LSTM model for stock price predictions was particularly difficult. Stock prices are inherently volatile and influenced by numerous unpredictable factors, making them one of the hardest data sets to model accurately. We had to experiment extensively with hyperparameters and model architecture to achieve meaningful results.
Training our finance chatbot was another hurdle. Ensuring the chatbot could understand and respond accurately to diverse finance-related questions required fine-tuning NLP models and training on a specialized dataset. Balancing the chatbot's accuracy and conversational flow presented its own set of complexities.
Using Streamlit to create an intuitive yet feature-rich interface also proved challenging. We needed to carefully design the user experience to ensure the platform remained accessible and engaging while handling complex financial data. Achieving the right balance between functionality and usability involved numerous iterations.
Finally, we faced the task of making sure every financial metric was correct and mathematically sound. This required extensive validation and error handling, as any discrepancies in data calculations or scoring would impact the overall analysis. Rigorous testing and verification were essential to maintain the platform’s accuracy and reliability.
Accomplishments that we're proud of
We’re proud of our achievements in developing a comprehensive financial analysis platform that offers clear and actionable insights to users. One of our standout accomplishments is our advanced LSTM model for stock price predictions. Given the unpredictable nature of financial markets, building a model that provides reliable and data-driven forecasts was a significant feat, empowering users to anticipate market movements with greater confidence.
Our Financial Health Dashboard features a meticulously crafted scoring system that translates complex financial metrics into an intuitive health score. By simplifying liquidity ratios, debt levels, and profitability measures into a cohesive analysis, we made it easy for users to gauge a company’s stability and make well-informed investment decisions.
Creating our finance chatbot was another accomplishment we're proud of. Using sophisticated NLP techniques, we trained the chatbot to provide instant, accurate answers to finance-related queries, making financial analysis accessible and engaging.
Finally, we successfully designed an intuitive and visually appealing interface using Streamlit. We transformed dense financial data into dynamic, interactive dashboards that are both user-friendly and insightful, offering a seamless experience that caters to both new users and financial experts.
What we learned
This project was a rich learning experience that expanded our understanding of financial analysis, machine learning, and user-centric design. We learned the intricacies of managing live financial data using the yFinance API, including strategies for ensuring data accuracy and handling real-time updates. This taught us the importance of efficient data processing and error handling to maintain reliability.
Developing our LSTM model for stock price forecasting challenged us to grapple with the unpredictable nature of financial markets. We gained valuable insights into time-series modeling, data normalization, and the complexities of tuning hyperparameters for optimal performance, all while learning to address overfitting and ensure robust predictions.
Our experience in creating a financial health scoring system deepened our understanding of financial metrics. We explored how to structure a points-based evaluation framework, carefully selecting and weighting each metric to create a clear, actionable assessment.
We also gained practical experience with Streamlit, which taught us how to design a clean, interactive user experience that balances complexity and usability. Training the finance chatbot enhanced our grasp of NLP, and we learned how to leverage AI to provide valuable, real-time assistance. Overall, we developed a holistic approach to building impactful financial solutions.
What's next for Finance101
Next for Finance101 is an ambitious roadmap focused on enhancing user experience, expanding predictive capabilities, and integrating more advanced financial tools. We plan to improve our LSTM model by incorporating additional features, such as sentiment analysis from news data, to make stock price predictions even more accurate and responsive to market conditions.
We also envision expanding the functionality of our finance chatbot to handle more complex financial queries, offering personalized investment insights and real-time portfolio analysis. This will involve training the chatbot with more comprehensive datasets and enhancing its conversational abilities using advanced NLP models. Another goal is to implement machine learning algorithms for portfolio optimization and risk assessment, helping users construct well-diversified and strategically balanced portfolios. Finally, we hope to release a mobile version of Finance101, ensuring that our platform is accessible and user-friendly across all devices. These advancements will solidify Finance101 as a comprehensive, cutting-edge financial analysis tool.
Log in or sign up for Devpost to join the conversation.