Review of Sentiment Analysis in Cryptocurrency Trading

Abstract

The rapid rise of cryptocurrencies has impacted the global socio-economic landscape, encouraging investors to seek income through crypto trading. Due to the market’s volatility and complex interdependencies, researchers have built various prediction models using machine learning, deep-learning, and sentiment-based hybrid algorithms. Notably, the DLCFS (Deep Learning Cryptocurrency Forecasting considering Sentiment) framework incorporates market features, trading volume, and sentiment from Reddit to improve price predictions for Bitcoin, Ethereum, and Litecoin—achieving high accuracy when compared to traditional machine learning models. Alongside forecasting, sentiment analysis plays an important role in understanding market trends and investor behavior. With growing user-generated content across different platforms like social media and news sites, extracting public sentiment through NLP has become essential. Recent works explore advanced models and datasets tailored to the unique linguistic features of crypto-related content, highlighting the need for robust and adaptive sentiment analysis techniques in this dynamic domain.