Topic time series analysis of microblogs

Abstract

Social media data tend to cluster around events and themes. Local newsworthy events, sports team victories or defeats, abnormal weather patterns and globally trending topics all influence the content of online discussion. The automated discovery of these underlying themes from corpora of text is of interest to numerous academic fields as well as to law enforcement organizations and commercial users. One useful class of tools to deal with such problems are topic models, which attempt to recover latent groups of word associations from the text. However, it is clear that these topics may also exhibit patterns in both time and space. The recovery of such patterns complements the analysis of the text itself and in many cases provides additional context. In this work we describe two methods for mining interesting spatio-temporal dynamics and relations among topics, one that compares the topic distributions as histograms in space and time and another that models topics over time as temporal or spatio-temporal Hawkes process with exponential trigger functions. Both methods may be used to discover topics with abnormal distributions in space and time. The second method also allows for self-exciting topics and can recover intertopic relationships (excitation or inhibition) in both time and space. We apply these methods to a geo-tagged Twitter dataset and provide analysis and discussion of the results.

Decoding the sentiment dynamics of online retailing customers: Time series analysis of social media

“…Surprisingly, the influence of issue salience in TV debates on Twitter activities was found to be limited or close to non-existent. Lai et al (2016) explored topic histograms and the emergence of topics over time to identify topics with an abnormal structure and recovered hidden interactions between the topics. Finally, to understand the influence of traditional media sources on tweets, Kim, Gonzenbach, Vargo, and Kim (2016) analysed how the issues and agendas of three media sources, TV advertisements, newspapers and Twitter, were correlated using autoregressive integrated moving average (ARIMA) time series.…”

Influence Dynamics Among Narratives: A Case Study of the Venezuelan Presidential Crisis

“…A couple of other examples that we mention here are are based on TPPs, which are used to model topic dynamics. Lai et al [12] propose using a marked Hawkes process to uncover intertopic relationships. Similarly, Mohler et al [14] propose a Hawkes-Binomial topic model to detect mutual influence between online Twitter activity and real-world events.…”

Machine learning for cardiac ultrasound time series data

“…7 In particular it is a dimensionality reduction technique that seeks to find lower paramertizations for high dimensionality data. In our analysis, Rank-2 NMF was implemented on each cardiac ultrasound video for the apical 4 view, the apical 2 view, the short axis view and the long axis view by using the following procedure.…”