Featured Work

StockSync

StockSync is a Flask-based stock market app that enables user authentication, wishlist management, and stock price prediction using machine learning.
Tech stacks
  • HTML,JS | Tailwind
  • yfinance | Flask
  • ML: Scikit-Learn
  • Database: MongoDB
Tech stacks
  • HTML,JS | Tailwind
  • yfinance | Flask
  • ML: Scikit-Learn
  • Database: MongoDB
Overview

StockSync is a stock price prediction model developed in May 2024, designed to provide accurate financial forecasts using machine learning techniques. The model achieved high accuracy with an MSE of 0.034, MAE of 0.123, R² score of 0.89, and EVS of 0.88. To enhance performance, a robust MongoDB data storage solution was implemented, improving application efficiency and user satisfaction by 50%. Additionally, data retrieval processes were optimized, increasing system reliability by 40% and reducing query response time by 60%. Built using Flask for the backend, MongoDB for database management, and yfinance and scikit-learn for data processing, StockSync ensures precise stock predictions while maintaining fast and efficient system performance.

Discover.

Highlights.
  • Flask-Based Web Application
  • Machine Learning Integration
  • MongoDB Database
  • Scalable & Secure
Features.
  • User Authentication
  • Wishlist Management
  • Stock Price Prediction
  • Interactive Dashboard
Approach

StockSync follows a structured approach to provide an efficient and user-friendly stock market application. The development process involves multiple key components, including user authentication, stock data retrieval, machine learning-based prediction, and an interactive web interface.

1. User Authentication & Management:To ensure security and personalized user experience, StockSync implements a user authentication system using Flask. Users can register, log in, and log out securely, with credentials stored in MongoDB. Passwords are hashed before storage to enhance security.
2. Stock Data Retrieval:The application fetches real-time stock data using the yfinance library. It retrieves historical data for the past six months to analyze trends and predict stock prices. The latest market data is displayed on the dashboard for users to make informed decisions.
3. Machine Learning-Based Prediction: Stock price prediction is implemented using Linear Regression from scikit-learn. The model is trained on historical stock prices, with dates converted into ordinal format as input features. The model then predicts the next day's closing price based on recent trends. Performance is evaluated using Mean Squared Error (MSE) and R² scores to ensure accuracy.
4. Wishlist Management: Users can add, view, and remove stocks from a personalized wishlist. This feature allows them to track their favorite stocks easily. The wishlist data is stored in MongoDB, ensuring quick retrieval and persistence across sessions.