Deeper Attention to Abusive User Content Moderation

John Pavlopoulos, Prodromos Malakasiotis, Ion Androutsopoulos


Abstract
Experimenting with a new dataset of 1.6M user comments from a news portal and an existing dataset of 115K Wikipedia talk page comments, we show that an RNN operating on word embeddings outpeforms the previous state of the art in moderation, which used logistic regression or an MLP classifier with character or word n-grams. We also compare against a CNN operating on word embeddings, and a word-list baseline. A novel, deep, classificationspecific attention mechanism improves the performance of the RNN further, and can also highlight suspicious words for free, without including highlighted words in the training data. We consider both fully automatic and semi-automatic moderation.
Anthology ID:
D17-1117
Volume:
Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing
Month:
September
Year:
2017
Address:
Copenhagen, Denmark
Editors:
Martha Palmer, Rebecca Hwa, Sebastian Riedel
Venue:
EMNLP
SIG:
SIGDAT
Publisher:
Association for Computational Linguistics
Note:
Pages:
1125–1135
Language:
URL:
https://aclanthology.org/D17-1117
DOI:
10.18653/v1/D17-1117
Bibkey:
Cite (ACL):
John Pavlopoulos, Prodromos Malakasiotis, and Ion Androutsopoulos. 2017. Deeper Attention to Abusive User Content Moderation. In Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, pages 1125–1135, Copenhagen, Denmark. Association for Computational Linguistics.
Cite (Informal):
Deeper Attention to Abusive User Content Moderation (Pavlopoulos et al., EMNLP 2017)
Copy Citation:
PDF:
https://preview.aclanthology.org/nschneid-patch-2/D17-1117.pdf
Video:
 https://preview.aclanthology.org/nschneid-patch-2/D17-1117.mp4