Cascade-based Echo Chamber Detection

Abstract

Despite echo chambers in social media have been under considerable scrutiny, general models for their detection and analysis are missing. In this work, we aim to ll this gap by proposing a probabilistic generative model that explains social media footprints-i.e., social network structure and propagations of information-through a set of latent communities, characterized by a degree of echochamber behavior and by an opinion polarity. Speci cally, echo chambers are modeled as communities that are permeable to pieces of information with similar ideological polarity, and impermeable to information of opposed leaning: this allows discriminating echo chambers from communities that lack a clear ideological alignment.To learn the model parameters we propose a scalable, stochastic adaptation of the Generalized Expectation Maximization algorithm, that optimizes the joint likelihood of observing social connections and information propagation. Experiments on synthetic data show that our algorithm is able to correctly reconstruct ground-truth latent communities with their degree of echo-chamber behavior and opinion polarity. Experiments on real-world data about polarized social and political debates, such as the Brexit referendum or the COVID-19 vaccine campaign, con rm the e ectiveness of our proposal in detecting echo chambers. Finally, we show how our model can improve accuracy in auxiliary predictive tasks, such as stance detection and prediction of future propagations.

Echo Chambers in Online Social Networks: A Systematic Literature Review

“…Content based [26]- [28], [30], [45], [46], [55] Information propagation based [28], [29], [53] Metric based [47], [51], [52], [56] Opinion dynamic based [44], [45] Polarization based [47], [49]- [51], [55] Topology based [28]- [30], [48], [52], [53] Additionally, given the powerful role of machine learning, especially deep learning techniques, in certain downstream tasks within echo chamber studies, we have highlighted the techniques used by existing studies for specific purposes, as shown in Table 8. Detailed discussions will be provided in Section V.…”

“…2 [27] Phenomena wherein online users exhibit selective exposure and ideological segregation on political issues. 3 [28] Groups of users exposed only to like-minded individuals tend to reinforce pre-existing opinions, a phenomenon scrutinized as a possible cause of increased polarization. 4 [29] The amplification of opinions through communication and repetition within a relatively closed social system.…”

“…In this review, we adopt the latter perspective Through our analysis, we found that the majority of existing studies in this review explicitly or implicitly view echo chambers as harmful, while the remaining studies do not provide a clear stance on the matter. The studies we reviewed highlighted that the primary effect and implication of echo chambers in online social networks is polarization and radicalization [28], [30], [31], [34], [39], [40], [42]- [44], [47], [49], [51], [52]. The prevailing viewpoint in these studies is that echo chambers are characterized by polarized groups.…”

“…Minici et al in [28] employed a hybrid approach to detect echo chambers by investigating the propagation of polarized information in a social network, drawing inspiration from community detection studies. They distinguished between two types of communities: echo chambers and social communities.…”

Moving Beyond Benchmarks and Competitions: Towards Addressing Social Media Challenges in an Educational Context

“…As there is no standard approach for the detection of echo chambers [84] we adopt commonly used ideas to this approach. We first apply language models for topic identification to the user's feed and the timeline posts of users connected to them in a one-hop neighborhood.…”