Comment toxicity detection via a multichannel convolutional bidirectional gated recurrent unit
Recently, toxicity identification has become the most serious problem in online communities and social networking sites. Therefore, an automatic toxic identification system needs to be developed for preventing and limiting users from these online environments. In this paper, we present a multichanne...
Saved in:
Main Authors: | , , , |
---|---|
Other Authors: | |
Format: | Article |
Language: | English |
Published: |
2022
|
Subjects: | |
Online Access: | https://hdl.handle.net/10356/160803 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
id |
sg-ntu-dr.10356-160803 |
---|---|
record_format |
dspace |
spelling |
sg-ntu-dr.10356-1608032022-08-03T02:25:19Z Comment toxicity detection via a multichannel convolutional bidirectional gated recurrent unit Kumar, J. Ashok Abirami, S. Trueman, Tina Esther Cambria, Erik School of Computer Science and Engineering Engineering::Computer science and engineering Multilabel Classification Multichannel Recently, toxicity identification has become the most serious problem in online communities and social networking sites. Therefore, an automatic toxic identification system needs to be developed for preventing and limiting users from these online environments. In this paper, we present a multichannel convolutional bidirectional gated recurrent unit (MCBiGRU) for detecting toxic comments in a multilabel environment. The proposed model generates word vectors using pre-trained word embeddings. Moreover, this hybrid model extracts local features with many filters and different kernel sizes to model input words with long term dependency. We then integrate multiple channels with a fully connected layer, normalization layer, and an output layer with a sigmoid activation function for predicting multilabel categories. The experimental results indicate that the proposed MCBiGRU model outperforms in terms of multilabel metrics. We thank the University Grants Commission (UGC), Government of India for supporting this work under the National Doctoral Fellowship. 2022-08-03T02:25:19Z 2022-08-03T02:25:19Z 2021 Journal Article Kumar, J. A., Abirami, S., Trueman, T. E. & Cambria, E. (2021). Comment toxicity detection via a multichannel convolutional bidirectional gated recurrent unit. Neurocomputing, 441, 272-278. https://dx.doi.org/10.1016/j.neucom.2021.02.023 0925-2312 https://hdl.handle.net/10356/160803 10.1016/j.neucom.2021.02.023 2-s2.0-85102621322 441 272 278 en Neurocomputing © 2021 Elsevier B.V. All rights reserved. |
institution |
Nanyang Technological University |
building |
NTU Library |
continent |
Asia |
country |
Singapore Singapore |
content_provider |
NTU Library |
collection |
DR-NTU |
language |
English |
topic |
Engineering::Computer science and engineering Multilabel Classification Multichannel |
spellingShingle |
Engineering::Computer science and engineering Multilabel Classification Multichannel Kumar, J. Ashok Abirami, S. Trueman, Tina Esther Cambria, Erik Comment toxicity detection via a multichannel convolutional bidirectional gated recurrent unit |
description |
Recently, toxicity identification has become the most serious problem in online communities and social networking sites. Therefore, an automatic toxic identification system needs to be developed for preventing and limiting users from these online environments. In this paper, we present a multichannel convolutional bidirectional gated recurrent unit (MCBiGRU) for detecting toxic comments in a multilabel environment. The proposed model generates word vectors using pre-trained word embeddings. Moreover, this hybrid model extracts local features with many filters and different kernel sizes to model input words with long term dependency. We then integrate multiple channels with a fully connected layer, normalization layer, and an output layer with a sigmoid activation function for predicting multilabel categories. The experimental results indicate that the proposed MCBiGRU model outperforms in terms of multilabel metrics. |
author2 |
School of Computer Science and Engineering |
author_facet |
School of Computer Science and Engineering Kumar, J. Ashok Abirami, S. Trueman, Tina Esther Cambria, Erik |
format |
Article |
author |
Kumar, J. Ashok Abirami, S. Trueman, Tina Esther Cambria, Erik |
author_sort |
Kumar, J. Ashok |
title |
Comment toxicity detection via a multichannel convolutional bidirectional gated recurrent unit |
title_short |
Comment toxicity detection via a multichannel convolutional bidirectional gated recurrent unit |
title_full |
Comment toxicity detection via a multichannel convolutional bidirectional gated recurrent unit |
title_fullStr |
Comment toxicity detection via a multichannel convolutional bidirectional gated recurrent unit |
title_full_unstemmed |
Comment toxicity detection via a multichannel convolutional bidirectional gated recurrent unit |
title_sort |
comment toxicity detection via a multichannel convolutional bidirectional gated recurrent unit |
publishDate |
2022 |
url |
https://hdl.handle.net/10356/160803 |
_version_ |
1743119605586460672 |