Skip to content
All library documents

Deep Learning for Crypto Price Classification with Trading and Social Signals

Article arXiv papers · Author: Marco Ortu et al.
Read the original

Summary

The study tests whether trading and social media indicators improve cryptocurrency price-change classification beyond technical indicators alone. It evaluates hourly and daily Bitcoin and Ethereum data from January 2017 to January 2021, comparing a restricted feature set of technical indicators with an expanded set that also includes trading and social signals. The models include multilayer perceptrons, convolutional networks, LSTM networks, and attention-based LSTMs. Grid search selects hyperparameters, and an advanced bootstrap method is used to account for variation in test samples.

For hourly predictions, the expanded feature set outperforms the technical-only version. Reported accuracy rises from 51–55% to 67–84% when trading indicators are added to the technical features. The excerpt does not provide corresponding daily results or detail how performance varies across models, assets, or market regimes. These findings concern classification in the study’s historical sample; accuracy alone does not establish profitability after trading costs.

Key ideas

  • The study compares technical-only features with technical, trading, and social media indicators.
  • It tests Bitcoin and Ethereum on hourly and daily observations from 2017 to 2021.
  • Four neural network approaches are evaluated, with grid search for hyperparameter selection.
  • Bootstrap analysis is used to account for variability in test samples.
  • On hourly data, reported accuracy improves from 51–55% to 67–84% when trading indicators are included.

Tags

CryptoMachine learningTechnical indicatorsSentimentStatistics
Full text
# On Technical Trading and Social Media Indicators in Cryptocurrencies' Price Classification Through Deep Learning


# On Technical Trading and Social Media Indicators in Cryptocurrencies' Price Classification Through Deep Learning









This work aims to analyse the predictability of price movements of cryptocurrencies on both hourly and daily data observed from January 2017 to January 2021, using deep learning algorithms. For our experiments, we used three sets of features: technical, trading and social media indicators, considering a restricted model of only technical indicators and an unrestricted model with technical, trading and social media indicators. We verified whether the consideration of trading and social media indicators, along with the classic technical variables (such as price's returns), leads to a significative improvement in the prediction of cryptocurrencies price's changes. We conducted the study on the two highest cryptocurrencies in volume and value (at the time of the study): Bitcoin and Ethereum. We implemented four different machine learning algorithms typically used in time-series classification problems: Multi Layers Perceptron (MLP), Convolutional Neural Network (CNN), Long Short Term Memory (LSTM) neural network and Attention Long Short Term Memory (ALSTM). We devised the experiments using the advanced bootstrap technique to consider the variance problem on test samples, which allowed us to evaluate a more reliable estimate of the model's performance. Furthermore, the Grid Search technique was used to find the best hyperparameters values for each implemented algorithm. The study shows that, based on the hourly frequency results, the unrestricted model outperforms the restricted one. The addition of the trading indicators to the classic technical indicators improves the accuracy of Bitcoin and Ethereum price's changes prediction, with an increase of accuracy from a range of 51-55% for the restricted model, to 67-84% for the unrestricted model.

Shown in full with attribution under the source's licence. Licence: abstract CC0

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.