Predicting Cryptocurrency Prices Using Historical Market Indicators and News 

Cryptocurrencies have gained significant attention in recent years, with their prices demonstrating high volatility. Accurately predicting cryptocurrency prices is of great interest to investors and traders. Previous studies have explored the use of historical market indicators, such as past prices, to forecast future cryptocurrency prices using machine learning techniques. However, the impact of news events on cryptocurrency prices cannot be overlooked. A notable example is when Elon Musk replaced Twitter’s bluebird icon with the Shiba Inu digital currency symbol, causing the price of Dogecoin to surge by over 30%. 

While some studies have investigated the effect of news sentiment on cryptocurrency prices independently, the results have been mixed, with some findings suggesting it is effective and others indicating otherwise. Moreover, the combined impact of news itself and technical indicators on cryptocurrency prices has not been extensively explored.  

To address this gap, we propose a project that aims to develop an effective model for predicting cryptocurrency prices by considering both historical technical indicators and news data. By leveraging advanced machine learning techniques such as Long Short-Term Memory (LSTM) and Transformers, we seek to capture the complex relationships between news topics, market sentiment, and price movements. The objective is to create a predictive model that outperforms existing approaches by incorporating a comprehensive set of features and employing advanced machine learning algorithms. 

Through this project, we aim to provide valuable insights into the factors influencing cryptocurrency prices and contribute to the development of more accurate and reliable price prediction models. The findings of this study could have significant implications for investors, traders, and researchers in the field of cryptocurrency market analysis.