This project was created as part of a bachelor's module (HTW Berlin) on AI applications and was awarded a final grade of 1.0. The main goal of the project was to implement a selected paper and verify its results. The specific paper implemented was: Stock Price Predictions with LSTM Neural Networks and Twitter Sentiment (found in the "project files" folder).
- Project Goal: Implement the mentioned paper and verify its results.
- Thematic Focus: The paper examines the prediction of stock prices using LSTM neural networks and sentiment analysis of Twitter data. Specifically, it analyzes whether the stock price of Apple can be predicted through sentiment analysis of Twitter posts.
- Project Structure: Since this project represents my first experience with artificial intelligence, its structure is unorthodox, and no significant results were achieved.
- Data Collection: The data was self-scraped in 2022 using SNScraper, an approach that is no longer applicable in 2024.
The main technologies and libraries used in this project are:
- Programming Language: Python
- Machine Learning Framework: PyTorch
- Data Analysis and Manipulation: Pandas, NumPy
- Web Scraping: SNScrape
- Sentiment Analysis: TextBlob
- Financial Data: yfinance
These technologies were utilized to gather data, analyze it, and train the LSTM model. The project provides a solid foundation for further research and experiments in stock price prediction using artificial intelligence and sentiment analysis.