Deep Learning-Based Cryptocurrency Sentiment Construction
20 Pages Posted: 8 Jan 2019
Date Written: December 10, 2018
We study investor sentiment on a non-classical asset, cryptocurrencies using a “crypto-specific lexicon” recently proposed in Chen et al. (2018) and statistical learning methods. We account for context-specific information and word similarity by learning word embeddings via neural network-based Word2Vec model. On top of pre-trained word vectors, we apply popular machine learning methods such as recursive neural networks for sentence-level classification and sentiment index construction. We perform this analysis on a novel dataset of 1220K messages related to 425 cryptocurrencies posted on a microblogging platform StockTwits during the period between March 2013 and May 2018. The constructed sentiment indices are value-relevant in terms of its return and volatility predictability for the cryptocurrency market index.
Keywords: sentiment analysis, lexicon, social media, word embedding, deep learning
JEL Classification: G41, G4, G12
Suggested Citation: Suggested Citation