Sensing Social Media Signals for Cryptocurrency News

@article{Beck2019SensingSM,
  title={Sensing Social Media Signals for Cryptocurrency News},
  author={Johannes Beck and Roberta Huang and David Lindner and Tian Guo and Zhang Ce and Dirk Helbing and Nino Antulov-Fantulin},
  journal={Companion Proceedings of The 2019 World Wide Web Conference},
  year={2019}
}
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 cryptocurrency news, which recently became of emerging interest to the general and financial audience. In order to track popular 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… Expand
Ask "Who", Not "What": Bitcoin Volatility Forecasting with Twitter Data
TLDR
This study focuses on volatility predictions for a relatively new asset class of cryptocurrencies (in particular, Bitcoin) using deep learning representations of public social media data from Twitter, finding statistical evidences for the hypotheses that temporal convolutional networks perform significantly better than both autoregressive and other deep learning-based models in the literature. Expand
Temporal mixture ensemble models for probabilistic forecasting of intraday cryptocurrency volume
TLDR
This work proposes a temporal mixture ensemble, capable of adaptively exploiting, for the forecasting, different sources of data and providing a volume point estimate, as well as its uncertainty, that performs slightly better than Gradient Boosting Machine while having a much clearer interpretability of the results. Expand
On the challenges of predicting microscopic dynamics of online conversations
TLDR
This work presents a generative model to predict the size and evolution of threaded conversations on social media by combining machine learning algorithms and shows that it is able to predict both macroscopic features of the final trees and near-future microscopic events with moderate accuracy. Expand
ESG2Risk: A Deep Learning Framework from ESG News to Stock Volatility Prediction
TLDR
Experimental evaluation of the proposed pipeline of ESG news extraction, news representations, and Bayesian inference of deep learning models demonstrates the superior predicting performance as well as the relation of high volatility prediction to stocks with potential high risk and low return. Expand
A Data Quality-Driven View of MLOps
TLDR
By performing a joint analysis of the impact of well-known data quality dimensions and the downstream machine learning process, it is shown that different components of a typical MLOps pipeline can be efficiently designed, providing both a technical and theoretical perspective. Expand
Ease.ML: A Lifecycle Management System for MLDev and MLOps
TLDR
This work presents Ease.ML, a lifecycle management system for machine learning that focuses on managing and automating the entire lifecycle itself, in terms of a probabilistic database model and its connection to information theory. Expand
Time-Varying Volatility in Bitcoin Market and Information Flow at Minute-Level Frequency
In this article, we analyze the time series of minute price returns on the Bitcoin market through the statistical models of the generalized autoregressive conditional heteroscedasticity (GARCH)Expand

References

SHOWING 1-10 OF 26 REFERENCES
Insight4News: Connecting News to Relevant Social Conversations
TLDR
The Insight4News system that connects news articles to social conversations, as echoed in microblogs such as Twitter, builds on the award winning Twitter topic detection approach and several machine learning components, to deliver news in a social context. Expand
The Pulse of News in Social Media: Forecasting Popularity
TLDR
This paper constructs a multi-dimensional feature space derived from properties of an article and evaluates the efficacy of these features to serve as predictors of online popularity and demonstrates that despite randomness in human behavior, it is possible to predict ranges of popularity on twitter with an overall 84% accuracy. Expand
Trading Strategies to Exploit Blog and News Sentiment
TLDR
A sentiment-based market-neutral trading strategy is given which gives consistently favorable returns with low volatility over a long period and several remarkable differences between news and blogs are identified. Expand
Listening to Chaotic Whispers: A Deep Learning Framework for News-oriented Stock Trend Prediction
TLDR
A Hybrid Attention Networks (HAN) is designed to predict the stock trend based on the sequence of recent related news, and the self-paced learning mechanism is applied to imitate the third principle. Expand
TiDeH: Time-Dependent Hawkes Process for Predicting Retweet Dynamics
TLDR
This paper model the Twitter system by Time-Dependent Hawkes process (TiDeH), which properly takes into account the circadian nature of the users and the aging of information, and develops a procedure for parameter optimization and for predicting the future profiles of retweet activity at different time resolutions. Expand
Linking online news and social media
TLDR
To combat query drift resulting from the large volume of text, either in the source news article itself or in social media utterances explicitly linked to it, a graph-based method for selecting discriminative terms is shown to help improve both effectiveness and efficiency. Expand
Predicting Socio-Economic Indicators using News Events
TLDR
This paper proposes a novel generative model of real-world events and employs it to extract events from a large corpus of news articles, and introduces the notion of an event class, which is an abstract grouping of similarly themed events. Expand
customers as advisors: the role of social media in financial markets
This paper investigates the extent to which peer-based advice transmitted through social media aects the stock market. We conduct textual analysis of articles published on Seeking Alpha, a popularExpand
Cohesiveness in Financial News and its Relation to Market Volatility
TLDR
A novel measure of collective behaviour based on financial news on the Web, the News Cohesiveness Index (NCI), is proposed and it is demonstrated that the index can be used as a financial market volatility indicator. Expand
Resampling Approaches to Improve News Importance Prediction
TLDR
The task of trying to forecast the number of times a news item will be tweeted, as a proxy for the importance assigned by its readers is addressed. Expand
...
1
2
3
...