Out of the Echo Chamber: Detecting Countering Debate Speeches

  title={Out of the Echo Chamber: Detecting Countering Debate Speeches},
  author={Matan Orbach and Yonatan Bilu and Assaf Toledo and Dan Lahav and Michal Jacovi and Ranit Aharonov and Noam Slonim},
An educated and informed consumption of media content has become a challenge in modern times. With the shift from traditional news outlets to social media and similar venues, a major concern is that readers are becoming encapsulated in “echo chambers” and may fall prey to fake news and disinformation, lacking easy access to dissenting views. We suggest a novel task aiming to alleviate some of these concerns – that of detecting articles that most effectively counter the arguments – and not just… 

Figures and Tables from this paper

Counter-TWIT: An Italian Corpus for Online Counterspeech in Ecological Contexts

The results highlight the fundamental role played by the context in this counterspeech detection task, confirming intuitions about the importance to collect tweets in their context of occurrence.

Conclusion-based Counter-Argument Generation

In real-world debates, the most common way to counter an argument is to reason against its main point, that is, its conclusion. Existing work on the automatic generation of natural language

Counter-Argument Generation by Attacking Weak Premises

It is hypothesized that identifying the argument’s weak premises is key to effective countering, and a pipeline approach is proposed that first assesses the premises’ strength and then generates a counter-argument undermining the weakest among them.

M-Arg: Multimodal Argument Mining Dataset for Political Debates with Audio and Transcripts

M-Arg is presented, a multimodal argument mining dataset with a corpus of US 2020 presidential debates, annotated through crowd-sourced annotations that allows models to be trained to extract arguments from natural dialogue such as debates using information like the intonation and rhythm of the speaker.

Argument Undermining: Counter-Argument Generation by Attacking Weak Premises

This work goes beyond previous research by exploring argument undermining, that is, countering an argument by attacking one of its premises, and proposes a pipeline approach that assesses the premises’ strength and then generates a counter-argument undermining the weakest among them.

Acoustic-Prosodic, Lexical and Demographic Cues to Persuasiveness in Competitive Debate Speeches

We analyze the acoustic-prosodic and lexical correlates of persuasiveness, taking into account speaker, judge and debate characteristics in a novel data set of 674 audio profiles, transcripts,

Knowledge-Enhanced Evidence Retrieval for Counterargument Generation

This work builds a system that, given a statement, retrieves counterevidence from diverse sources on the Web and presents a knowledge-enhanced NLI model that aims to handle causality and example-based inference by incorporating knowledge graphs.

Topic Ontologies for Arguments

This paper contributes the first comprehensive survey of topic coverage, assessing 45 argument corpora and shows that the corpora topics— which are mostly those frequently discussed in public online fora—are covered well by the sources.

Tool to Retrieve Less-Filtered Information from the Internet

This work implements an approach based on the mechanism of a metasearch engine to present less-filtered information to users, and develops a practical application named MosaicSearch to select search results from diversified categories of sources collected from multiple search engines.

A Survey on Echo Chambers on Social Media: Description, Detection and Mitigation

This survey aims to examine the echo chamber phenomenon on social media from a social computing perspective and provide a blueprint for possible solutions and shows the mechanisms, both algorithmic and psychological, that lead to the formation of echo chambers.



Towards Effective Rebuttal: Listening Comprehension Using Corpus-Wide Claim Mining

A large dataset of 400 speeches in English discussing 200 controversial topics, mined claims for each topic, and asked annotators to identify the mined claims mentioned in each speech shows that in the vast majority of speeches debaters indeed make use of such claims.

Political Discourse on Social Media: Echo Chambers, Gatekeepers, and the Price of Bipartisanship

It is found that Twitter users are, to a large degree, exposed to political opinions that agree with their own, and users who try to bridge the echo chambers have to pay a »price of bipartisanship» in terms of their network centrality and content appreciation.

A Dataset of General-Purpose Rebuttal

The general rebuttals described here overcome the need for topic-specific arguments to be provided, by proving to be applicable for a large set of topics, which allows creating responses beyond the scope of topics for which specific arguments are available.

A Corpus for Research on Deliberation and Debate

This paper presents the Internet Argument Corpus, a set of 390,704 posts in 11,800 discussions extracted from the online debate site 4forums.com, and discusses the relationship between discourse marker pragmatics, agreement, emotionality, and sarcasm in the IAC corpus.

I Couldn’t Agree More: The Role of Conversational Structure in Agreement and Disagreement Detection in Online Discussions

This paper shows how different aspects of conversational structure can be used to detect agreement and disagreement in discussion forums, and exploits information about meta-thread structure and accommodation between participants to outperform the same system trained on prior existing in-domain smaller annotated datasets.

A Recorded Debating Dataset

This paper describes an audio and textual dataset of debating speeches, a first-of-a-kind resource for the growing research field of computational argumentation and debating technologies, and releases speeches on various controversial topics in 5 formats corresponding to the different stages in the production of the data.

Joint Models of Disagreement and Stance in Online Debate

This work comprehensively evaluates the possible modeling choices on eight topics across two online debate corpora and introduces a scalable unified probabilistic modeling framework for stance classification models that are collective, reason about disagreement, and can model stance at either the author level or at the post level.

Towards Detecting Counter-considerations in Text

A classification approach to classifying counter-considerations in monologue text is presented and applied to two different corpora: a selection of very short argumentative texts produced in a text generation experiment, and a set of newspaper commentaries.

A Sociolinguistic Study of Online Echo Chambers on Twitter

  • Nikita DusejaHarsh Jhamtani
  • Computer Science
    Proceedings of the Third Workshop on Natural Language Processing and Computational Social Science
  • 2019
This paper carries out a comparative analysis of tweet structure, lexical choices, and focus issues, and provides possible explanations for the results of a sociolinguistic study by comparing the tweets of users in echo chambers with the emails of users not in echo chamber with similar levels of polarity on a broad topic.

Listening Comprehension over Argumentative Content

This paper presents a task for machine listening comprehension in the argumentation domain and a corresponding dataset in English, and applied baseline methods addressing the task, to be used as a benchmark for future work over this dataset.