Kobra Stocks

KobraStocks is a web-based application designed to predict the movement of stock prices. Users can enter a stock ticker in the search bar to view specific graphs, data, and predictions of a stock.

Comment

KobraStocks

KU Hackathon 2024 Team Members

  • Cowell
  • Sliffe
  • Gillund

Introduction

KobraStocks is an open-source web-based application designed to predict the movement of stock prices. Users can input a stock ticker in the search bar to access graphs, data, and predictions. These tools help determine whether the stock price is likely to move up or down, alongside the accuracy of the model's predictions.

Features

  • Stock Search: Users can enter the stock ticker symbol to access the stock's page, which includes financial data and predictions. Flask dynamically renders the template with data for the searched stock.
  • Selectable Indicators: Users have the option to select various financial indicators such as MACD, RSI, and MA, which help in creating a personalized machine learning model.
  • Visualization: Utilizing the Python library Plotly, we generate detailed graphs that represent the historical data of the stock's price along with selected indicators. The graph is scalable, allowing users to focus on specific days or to view up to 5 years of data. A bar graph displaying the stock's trade volume is also included below the main graph.
  • Data Insights: Provides key data points related to the stock's performance, such as the High, Low, Open, and Close prices from the previous day.
  • Predictions: Leverages primary data points and Scikit-Learn's Random Forest Classifier to predict whether a stock's price will increase or decrease over the next day, week, or month. We assess the model's accuracy by comparing its current predictions with past predictions, achieving up to 80% accuracy weekly.

Version Control

  • Platforms Used: GitHub and Azure DevOps
    • These platforms help manage our codebase, ensuring smooth collaboration and effective tracking of changes across the project.

Technologies Used

  • Frontend: HTML, CSS, JavaScript
  • Backend: Python Flask
  • APIs: Yahoo Finance
  • Machine Learning: Scikit-Learn (Random Forest Classifier)
  • Cloud: Azure

Built With

Share this project:

Updates