TY - GEN
T1 - Pinpointing Fine-Grained Relationships between Hateful Tweets and Replies
AU - Albanyan, Abdullah
AU - Blanco, Eduardo
N1 - Publisher Copyright:
Copyright © 2022, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved.
PY - 2022/6/30
Y1 - 2022/6/30
N2 - Recent studies in the hate and counter hate domain have provided the grounds for investigating how to detect this pervasive content in social media. These studies mostly work with synthetic replies to hateful content written by annotators on demand rather than replies written by real users. We argue that working with naturally occurring replies to hateful content is key to study the problem. Building on this motivation, we create a corpus of 5,652 hateful tweets and replies. We analyze their fine-grained relationships by indicating whether the reply (a) is hate or counter hate speech, (b) provides a justification, (c) attacks the author of the tweet, and (d) adds additional hate. We also present linguistic insights into the language people use depending on these fine-grained relationships. Experimental results show improvements (a) taking into account the hateful tweet in addition to the reply and (b) pretraining with related tasks.
AB - Recent studies in the hate and counter hate domain have provided the grounds for investigating how to detect this pervasive content in social media. These studies mostly work with synthetic replies to hateful content written by annotators on demand rather than replies written by real users. We argue that working with naturally occurring replies to hateful content is key to study the problem. Building on this motivation, we create a corpus of 5,652 hateful tweets and replies. We analyze their fine-grained relationships by indicating whether the reply (a) is hate or counter hate speech, (b) provides a justification, (c) attacks the author of the tweet, and (d) adds additional hate. We also present linguistic insights into the language people use depending on these fine-grained relationships. Experimental results show improvements (a) taking into account the hateful tweet in addition to the reply and (b) pretraining with related tasks.
UR - http://www.scopus.com/inward/record.url?scp=85147542537&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85147542537&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:85147542537
T3 - Proceedings of the 36th AAAI Conference on Artificial Intelligence, AAAI 2022
SP - 10418
EP - 10426
BT - AAAI-22 Technical Tracks 10
PB - Association for the Advancement of Artificial Intelligence
T2 - 36th AAAI Conference on Artificial Intelligence, AAAI 2022
Y2 - 22 February 2022 through 1 March 2022
ER -