esramelikecakir/integration-of-machine-learning-and-social-media-data-for-stock-price-prediction

This project delves into the intriguing realm of how thoughts and information shared on platforms like Twitter wield influence over financial markets.

Integration-of-Machine-Learning-and-Social-Media-Data-for-Stock-Price-Prediction

In today's dynamic world, where millions of posts flood social media daily, an unstoppable wave of interaction engulfs people in every corner of their lives. This project delves into the intriguing realm of how thoughts and information shared on platforms like Twitter wield influence over financial markets. The analysis hones in on unraveling the intricate ways in which social media content can mold economic preferences and, in turn, steer the tides of stock prices.

Project Objectives

• Uncover the impact of Twitter posts on financial markets.
• Leverage the BERT model for sentiment analysis, decoding the emotional undertones of social media posts.
• Make sense of the relationship between social media shares and stock market transactions using statistical and data science methodologies.
• Enhance communication of statistical insights through vivid analyses with tools such as Matplotlib, Transformers, and Torch.

This project utilizes data sourced from Kaggle, emphasizing a comprehensive exploration of the relationship between social media and financial markets. Specifically, a deep dive into the impact of Twitter messages on Tesla stock has been undertaken.

In this endeavor, the project harnesses popular and versatile libraries such as Matplotlib, Transformers, and Torch for data analysis and visualization. One of the prediction methods employed is the MLPRegressor, which, when fed with financial indicators, forecasts stock price movements.

The computer programs employed here are of universal accessibility, particularly user-friendly and powerful tools like sklearn and pandas, employed for data analysis and visualization. This study not only provides fundamental insights but also stands as a visual and analytical spectacle, underlining the pervasive influence of social media on financial markets.

In conclusion, this thesis serves as an initiation point for understanding the intriguing relationship between social media messages and stock prices, aspiring to lay a robust foundation for deeper and more comprehensive research endeavors in the future.

In this link, you can find datas that we used in this project : https://www.kaggle.com/code/shreytandel19/stock-prediction-based-on-tweet-sentiment-analysis/input