Data-centered persuasion: Nudging user's prosocial behavior and designing social innovation

Shin, Youngsoo; Kim, Jin-Woo

doi:10.1016/j.chb.2017.11.009

“…Use of the r-word is unacceptable as it demeans and insults people with disabilities. ing countermeasure is counterspeech-a response that provides non-negative feedback through factbound arguments and broader perspectives to mitigate hate speech and fostering a more harmonious conversation in social platforms (Schieb and Preuss, 2016;Munger, 2017;Mathew et al, 2018;Shin and Kim, 2018). Counterspeech as a measure to combat abusive language online is also promoted in active campaigns such as "Get The Trolls Out".…”

Section: Not Relevantmentioning

confidence: 99%

Generate, Prune, Select: A Pipeline for Counterspeech Generation against Online Hate Speech

Zhu¹,

Bhat²

2021

Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021

View full text Add to dashboard Cite

Warning: this paper contains content that may be offensive or upsetting.Countermeasures to effectively fight the ever increasing hate speech online without blocking freedom of speech is of great social interest. Natural Language Generation (NLG), is uniquely capable of developing scalable solutions. However, off-the-shelf NLG methods are primarily sequence-to-sequence neural models and they are limited in that they generate commonplace, repetitive and safe responses regardless of the hate speech (e.g., "Please refrain from using such language.") or irrelevant responses, making them ineffective for de-escalating hateful conversations. In this paper, we design a three-module pipeline approach to effectively improve the diversity and relevance. Our proposed pipeline first generates various counterspeech candidates by a generative model to promote diversity, then filters the ungrammatical ones using a BERT model, and finally selects the most relevant counterspeech response using a novel retrievalbased method. Extensive Experiments on three representative datasets demonstrate the efficacy of our approach in generating diverse and relevant counterspeech.

show abstract

“…Use of the r-word is unacceptable as it demeans and insults people with disabilities. ing countermeasure is counterspeech-a response that provides non-negative feedback through factbound arguments and broader perspectives to mitigate hate speech and fostering a more harmonious conversation in social platforms (Schieb and Preuss, 2016;Munger, 2017;Mathew et al, 2018;Shin and Kim, 2018). Counterspeech as a measure to combat abusive language online is also promoted in active campaigns such as "Get The Trolls Out".…”

Section: Not Relevantmentioning

confidence: 99%

Generate, Prune, Select: A Pipeline for Counterspeech Generation against Online Hate Speech

Zhu

¹

,

Bhat

²

2021

Preprint

View full text Add to dashboard Cite

Warning: this paper contains content that may be offensive or upsetting.Countermeasures to effectively fight the ever increasing hate speech online without blocking freedom of speech is of great social interest. Natural Language Generation (NLG), is uniquely capable of developing scalable solutions. However, off-the-shelf NLG methods are primarily sequence-to-sequence neural models and they are limited in that they generate commonplace, repetitive and safe responses regardless of the hate speech (e.g., "Please refrain from using such language.") or irrelevant responses, making them ineffective for de-escalating hateful conversations. In this paper, we design a three-module pipeline approach to effectively improve the diversity and relevance. Our proposed pipeline first generates various counterspeech candidates by a generative model to promote diversity, then filters the ungrammatical ones using a BERT model, and finally selects the most relevant counterspeech response using a novel retrievalbased method. Extensive Experiments on three representative datasets demonstrate the efficacy of our approach in generating diverse and relevant counterspeech.

show abstract

“…Despite the presence of control mechanisms within networked technology, part of the rhetorical appeals used to encourage the adoption of networked technology is to disguise that relationship under the guise of "individual agency," "empowerment," and a "culture of health" (Teston, 2016). In human-computer interaction research, the arguments for behavioral change embedded in persuasive technology are described benignly as "nudges" (Fogg, 2002;Shin & Kim, 2018). These framings of the relationship of the individual to the larger network locate the locus of control, with attending agency and responsibility for one's behavior, internally.…”

Section: Locus Of Controlmentioning

confidence: 99%

Rhetorical Implications of Contact Tracing Mobile Applications: An Examination of Big Data’s Work on the Body

Lanius

¹

2021

Poroi

0

View full text Add to dashboard Cite

For nearly a decade, big data has been hyped as an amazing new technology that will benefit corporations and consumers alike. By promising customized knowledge at an accelerated pace, big data technologies have slowly saturated the digital systems American consumers use to live, work, and play. Yet have the promised benefits materialized? An examination of the proposed contact tracing applications in response to the novel coronavirus alongside existing wearable technologies reveal that our trust and vulnerability, opening our bodies to be sensed by these networked systems, is a fraught rhetorical activity, but not because an omniscient system now sees us and cares for us in our time of grave need. Rather, the opaque system misunderstands our embodied rhetorical actions, is incapable of moving the American polis, and cannot generate the promised collective action.

show abstract

Data-centered persuasion: Nudging user's prosocial behavior and designing social innovation

Cited by 21 publications

References 28 publications

Generate, Prune, Select: A Pipeline for Counterspeech Generation against Online Hate Speech

Generate, Prune, Select: A Pipeline for Counterspeech Generation against Online Hate Speech

Generate, Prune, Select: A Pipeline for Counterspeech Generation against Online Hate Speech

Rhetorical Implications of Contact Tracing Mobile Applications: An Examination of Big Data’s Work on the Body

Contact Info

Product

Resources

About