Sensing Social Media Signals for Cryptocurrency News

Abstract

The ability to track and monitor relevant and important news in real-time is of crucial interest in multiple industrial sectors. In this work, we focus on the set of cryptocurrency news, which recently became of emerging interest to the general and financial audience. In order to track relevant news in real-time, we (i) match news from the web with tweets from social media, (ii) track their intraday tweet activity and (iii) explore different machine learning models for predicting the number of article mentions on Twitter within the first 24 hours after its publication. We compare several machine learning models, such as linear extrapolation, linear and random forest autoregressive models, and a sequence-to-sequence neural network. We find that the random forest autoregressive model behaves comparably to more complex models in the majority of tasks. * Contributed equally to the first authorship † anino@ethz.ch

Community impact on a cryptocurrency: Twitter comparison example between Dogecoin and Litecoin

“…Our study has thus proven the interest of behavioral economics applied to the cryptocurrency world like others before it [31,32,33]. However, many things remain to be improved: the application of this modeling to cryptocurrencies operating under other consensus mechanisms (Proof of Stake (PoS), Delegated Proof of Stake (DPoS), Practical Byzantine Fault Tolerance (PBFT)...), the application to cryptocurrencies with a voting consensus (for such consensus, the formulation of the problem would be more meaningful, and from the point of view of applications to commodity backed cryptocurrency their study would be more meaningful), creation of an online bot to predict cryptocurrencies future prices… Tables Table 1. Augmented Dickey-Fuller test results for the three types of linear regression models…”

Time-Varying Volatility in Bitcoin Market and Information Flow at Minute-Level Frequency

“…The price returns were calculated using two different definitions, VWAP and mid-quote, to account for possible market-microstructure noise. Another reason, why we have concentrated on the Bitcoin, was the availability of Twitter-related data [36]. We have used the social media signals from Twitter, trading volume and bid-ask spread from the Bitcoin market as a proxy for information flow together with the GARCH family of [53] processes to quantify the prediction power for the price volatility.…”

“…Three different datasets for external signals were available as the external information proxy-a time series of the number of tweets mentioning cryptocurrency-related news [36], a time series of Bitcoin trade volumes from Bitfinex market, and a time series of Bitcoin bid-ask spread, created as a time series of absolute differences between the maximum bid and the minimum ask price at every recorded instant, also from Bitfinex market. The data are collected on a second level and shown in Figures 2A-C, with the descriptive statistics in Figure 2D.…”

Ask "Who", Not "What": Bitcoin Volatility Forecasting with Twitter Data

“…Therefore, cryptocurrency markets have been studied from various alternative perspectives, ranging from volatility and volume forecasting using standard econometric models [33][34][35], to employing tools from systems dynamics [36][37][38]. Various studies analysed the effect of external data from social media, news, search queries, sentiment, comments, replies on forums, and blockchain [13,[39][40][41]. Typically, sentiment analysis is performed and combined with econometric models such as GARCH [7] or deep learning models such as CNN [4].…”