Loading [MathJax]/extensions/tex2jax.js

Why do we Hate Migrants? A Double Machine Learning-based Approach

Research output: Chapter in book/report/conference proceedingConference contributionResearchpeer review

Authors

Research Organisations

Details

Original languageEnglish
Title of host publicationProceedings of the 34th ACM Conference on Hypertext and Social Media
Number of pages10
ISBN (electronic)9798400702327
Publication statusPublished - 5 Sept 2023

Abstract

AI-based NLP literature has explored antipathy toward the marginalized section of society, such as migrants, and their social acceptance. Broadly, extant literature has conceptualized this as an online hate speech detection task and employed predictive ML models. However, a crucial omission in this literature is the genesis (or causality) of online hate, i.e., why do we hate migrants? Drawing insights from social science literature, we have identified three antecedents of online hate: Cultural, Economic, and Security concerns. Subsequently, we probe -which of these concerns triggers higher toxicity on online platforms? Initially, we consider OLS-based regression analysis and SHAP framework to identify the predictors of toxicity, and subsequently, we use Double Machine Learning (DML)-based casual analysis to investigate whether good predictors of toxicity are also causally significant. We find that the causal effect of Cultural concerns on toxicity is higher than Security and Economic concerns.

Keywords

    Causality, Double machine Learning, Online Hate, Toxicity

ASJC Scopus subject areas

Cite this

Why do we Hate Migrants? A Double Machine Learning-based Approach. / Khatua, Aparup; Nejdl, Wolfgang.
Proceedings of the 34th ACM Conference on Hypertext and Social Media. 2023. 35.

Research output: Chapter in book/report/conference proceedingConference contributionResearchpeer review

Khatua, A & Nejdl, W 2023, Why do we Hate Migrants? A Double Machine Learning-based Approach. in Proceedings of the 34th ACM Conference on Hypertext and Social Media., 35. https://doi.org/10.1145/3603163.3609040
Khatua, A., & Nejdl, W. (2023). Why do we Hate Migrants? A Double Machine Learning-based Approach. In Proceedings of the 34th ACM Conference on Hypertext and Social Media Article 35 https://doi.org/10.1145/3603163.3609040
Khatua A, Nejdl W. Why do we Hate Migrants? A Double Machine Learning-based Approach. In Proceedings of the 34th ACM Conference on Hypertext and Social Media. 2023. 35 doi: 10.1145/3603163.3609040
Khatua, Aparup ; Nejdl, Wolfgang. / Why do we Hate Migrants? A Double Machine Learning-based Approach. Proceedings of the 34th ACM Conference on Hypertext and Social Media. 2023.
Download
@inproceedings{6b020e0f4f5a4a84b43b475ea977ce94,
title = "Why do we Hate Migrants?: A Double Machine Learning-based Approach",
abstract = "AI-based NLP literature has explored antipathy toward the marginalized section of society, such as migrants, and their social acceptance. Broadly, extant literature has conceptualized this as an online hate speech detection task and employed predictive ML models. However, a crucial omission in this literature is the genesis (or causality) of online hate, i.e., why do we hate migrants? Drawing insights from social science literature, we have identified three antecedents of online hate: Cultural, Economic, and Security concerns. Subsequently, we probe -which of these concerns triggers higher toxicity on online platforms? Initially, we consider OLS-based regression analysis and SHAP framework to identify the predictors of toxicity, and subsequently, we use Double Machine Learning (DML)-based casual analysis to investigate whether good predictors of toxicity are also causally significant. We find that the causal effect of Cultural concerns on toxicity is higher than Security and Economic concerns.",
keywords = "Causality, Double machine Learning, Online Hate, Toxicity",
author = "Aparup Khatua and Wolfgang Nejdl",
note = "Funding Information: Funding for this paper was, in part, provided by the Federal Ministry of Education and Research (BMBF), Germany under the project LeibnizKILabor with grant No. 01DD20003.",
year = "2023",
month = sep,
day = "5",
doi = "10.1145/3603163.3609040",
language = "English",
isbn = "979-8-4007-0232-7",
booktitle = "Proceedings of the 34th ACM Conference on Hypertext and Social Media",

}

Download

TY - GEN

T1 - Why do we Hate Migrants?

T2 - A Double Machine Learning-based Approach

AU - Khatua, Aparup

AU - Nejdl, Wolfgang

N1 - Funding Information: Funding for this paper was, in part, provided by the Federal Ministry of Education and Research (BMBF), Germany under the project LeibnizKILabor with grant No. 01DD20003.

PY - 2023/9/5

Y1 - 2023/9/5

N2 - AI-based NLP literature has explored antipathy toward the marginalized section of society, such as migrants, and their social acceptance. Broadly, extant literature has conceptualized this as an online hate speech detection task and employed predictive ML models. However, a crucial omission in this literature is the genesis (or causality) of online hate, i.e., why do we hate migrants? Drawing insights from social science literature, we have identified three antecedents of online hate: Cultural, Economic, and Security concerns. Subsequently, we probe -which of these concerns triggers higher toxicity on online platforms? Initially, we consider OLS-based regression analysis and SHAP framework to identify the predictors of toxicity, and subsequently, we use Double Machine Learning (DML)-based casual analysis to investigate whether good predictors of toxicity are also causally significant. We find that the causal effect of Cultural concerns on toxicity is higher than Security and Economic concerns.

AB - AI-based NLP literature has explored antipathy toward the marginalized section of society, such as migrants, and their social acceptance. Broadly, extant literature has conceptualized this as an online hate speech detection task and employed predictive ML models. However, a crucial omission in this literature is the genesis (or causality) of online hate, i.e., why do we hate migrants? Drawing insights from social science literature, we have identified three antecedents of online hate: Cultural, Economic, and Security concerns. Subsequently, we probe -which of these concerns triggers higher toxicity on online platforms? Initially, we consider OLS-based regression analysis and SHAP framework to identify the predictors of toxicity, and subsequently, we use Double Machine Learning (DML)-based casual analysis to investigate whether good predictors of toxicity are also causally significant. We find that the causal effect of Cultural concerns on toxicity is higher than Security and Economic concerns.

KW - Causality

KW - Double machine Learning

KW - Online Hate

KW - Toxicity

UR - http://www.scopus.com/inward/record.url?scp=85174318700&partnerID=8YFLogxK

U2 - 10.1145/3603163.3609040

DO - 10.1145/3603163.3609040

M3 - Conference contribution

SN - 979-8-4007-0232-7

BT - Proceedings of the 34th ACM Conference on Hypertext and Social Media

ER -

By the same author(s)