Filters








3,603 Hits in 4.1 sec

Neural Character-based Composition Models for Abuse Detection [article]

Pushkar Mishra, Helen Yannakoudakis, Ekaterina Shutova
2018 arXiv   pre-print
The current state of the art approaches to abusive language detection, based on recurrent neural networks, do not explicitly address this problem and resort to a generic OOV (out of vocabulary) embedding  ...  In this paper, we address this problem by designing a model that can compose embeddings for unseen words.  ...  Acknowledgements Special thanks to the anonymous reviewers for their valuable comments and suggestions.  ... 
arXiv:1809.00378v1 fatcat:eetftlviw5hk7myihjixevgbha

Neural Character-based Composition Models for Abuse Detection

Pushkar Mishra, Helen Yannakoudakis, Ekaterina Shutova
2018 Proceedings of the 2nd Workshop on Abusive Language Online (ALW2)  
Acknowledgements Special thanks to the anonymous reviewers for their valuable comments and suggestions.  ...  To the best of our knowledge, this is the first work to use character-based word composition models for abuse detection.  ...  Taking inspiration from the effectiveness of character-level features in abuse detection, we address this issue by having a character-based word composition model that can compose embeddings for unseen  ... 
doi:10.18653/v1/w18-5101 dblp:conf/acl-alw/MishraYS18 fatcat:y7mbghaq6japjkuojozdhwhtjm

Birds of a Feather Flock Together: Generating Pornographic and Gambling Domain Names Based on Character Composition Similarity

Yanan Cheng, Hao Jiang, Zhaoxin Zhang, Yuejin Du, Tingting Chai, Yuanlong Cao
2022 Wireless Communications and Mobile Computing  
Additionally, we develop a two-layer detection system for pornography and gambling domains using fastText and CNN models to obtain an abusive domain dataset for model training and validation.  ...  Therefore, this study combines the ideas of text similarity and text generation and proposes an abusive domain generation model based on GRU for rapidly generating new abusive domain names from known ones  ...  [11] established a model for detecting domain generation algorithm (DGA) domains using recurrent neural networks.  ... 
doi:10.1155/2022/4408987 fatcat:j5cehl5webbgld6vroclzbp4l4

UTFPR at SemEval-2019 Task 5: Hate Speech Identification with Recurrent Neural Networks [article]

Gustavo Henrique Paetzold, Shervin Malmasi, Marcos Zampieri
2019 arXiv   pre-print
We approach the task using a system based on minimalistic compositional Recurrent Neural Networks (RNN).  ...  We tested our approach on the SemEval-2019 Task 5: Multilingual Detection of Hate Speech Against Immigrants and Women in Twitter (HatEval) shared task dataset.  ...  Acknowledgements We would like to thank the organizers of the HatEval shared task for providing participants with this dataset and for organizing this interesting shared task.  ... 
arXiv:1904.07839v1 fatcat:fppqqovxtjertmomrozfjqzfxu

UTFPR at SemEval-2019 Task 5: Hate Speech Identification with Recurrent Neural Networks

Gustavo Henrique Paetzold, Marcos Zampieri, Shervin Malmasi
2019 Proceedings of the 13th International Workshop on Semantic Evaluation  
We approach the task using a system based on minimalistic compositional Recurrent Neural Networks (RNN).  ...  We tested our approach on the SemEval-2019 Task 5: Multilingual Detection of Hate Speech Against Immigrants and Women in Twitter (HatEval) shared task dataset.  ...  Acknowledgements We would like to thank the organizers of the HatEval shared task for providing participants with this dataset and for organizing this interesting shared task.  ... 
doi:10.18653/v1/s19-2093 dblp:conf/semeval/PaetzoldZM19 fatcat:sgdhywlnane3hcig3jnynac6fq

Tackling Online Abuse: A Survey of Automated Abuse Detection Methods [article]

Pushkar Mishra, Helen Yannakoudakis, Ekaterina Shutova
2020 arXiv   pre-print
We describe the existing datasets and review the computational approaches to abuse detection, analyzing their strengths and limitations.  ...  Consequently, over the past few years, there has been a substantial research effort towards automated abuse detection in the field of natural language processing (NLP).  ...  Neural networks based abuse detection Advancements in computational capabilities have led researchers to explore methods for abuse detection that rely on neural architectures.  ... 
arXiv:1908.06024v2 fatcat:53nwpcbspjdbrnyplzmsyjxmpu

Detect All Abuse! Toward Universal Abusive Language Detection Models [article]

Kunze Wang, Dong Lu, Soyeon Caren Han, Siqu Long, Josiah Poon
2020 arXiv   pre-print
Online abusive language detection (ALD) has become a societal issue of increasing importance in recent years.  ...  Our generic framework covers multi-aspect abusive language embeddings that represent the target and content aspects of abusive language and applies a textual graph embedding that analyses the user's linguistic  ...  Multi-Features with RNN (MFR): MFR (Mehdad and Tetreault, 2016) used a hybrid character-based and word-based Recurrent Neural Network (RNN) model to detect abusive language.  ... 
arXiv:2010.03776v2 fatcat:e4xkjgg7qrfeposymvhvrazxci

A Multitask Learning Framework for Abuse Detection and Emotion Classification

Yucheng Huang, Rui Song, Fausto Giunchiglia, Hao Xu
2022 Algorithms  
Then, we used two different decoders for emotion classification and abuse detection, respectively.  ...  pretrained language model.  ...  In addition, we also implemented a word-level CNN and character-level CNN for comparison. • Multifeatures with RNN [49] , a hybrid character-based and word-based recursive neural network (RNN) model was  ... 
doi:10.3390/a15040116 fatcat:huu43b2hxfewxklwcbngwuan3e

Learning Representations for Detecting Abusive Language

Magnus Sahlgren, Tim Isbister, Fredrik Olsson
2018 Proceedings of the 2nd Workshop on Abusive Language Online (ALW2)  
This paper discusses the question whether it is possible to learn a generic representation that is useful for detecting various types of abusive language.  ...  We compare the learned representation with two standard approaches; one based on lexica, and one based on dataspecific n-grams.  ...  This is not the case for the language model, which operates on the character sequences, and therefore will produce different compositional representations for these two sequences.  ... 
doi:10.18653/v1/w18-5115 dblp:conf/acl-alw/SahlgrenIO18 fatcat:gycolejzrjf7vpc4uo6a2qxtry

Detecting Online Hate Speech Using Context Aware Models [article]

Lei Gao, Ruihong Huang
2018 arXiv   pre-print
Then we propose two types of hate speech detection models that incorporate context information, a logistic regression model with context features and a neural network model with learning components for  ...  Context accompanying a hate speech text is useful for identifying hate speech, which however has been largely overlooked in existing datasets and hate speech detection models.  ...  In addition, we presented two types of models, feature-based logistic regression models and neural network models, in order to incorporate context information for improving hate speech detection performance  ... 
arXiv:1710.07395v2 fatcat:lurzf7r47zglfihncprxvbyy7e

Detecting Online Hate Speech Using Context Aware Models

Lei Gao, Ruihong Huang
2017 RANLP 2017 - Recent Advances in Natural Language Processing Meet Deep Learning  
Then we propose two types of hate speech detection models that incorporate context information, a logistic regression model with context features and a neural network model with learning components for  ...  Context accompanying a hate speech text is useful for identifying hate speech, which however has been largely overlooked in existing datasets and hate speech detection models.  ...  In addition, we presented two types of models, feature based logistic regression models and neural network models, in order to incorporate context information for improving hate speech detection performance  ... 
doi:10.26615/978-954-452-049-6_036 dblp:conf/ranlp/GaoH17 fatcat:syu7ktngdvashmaaqefti63vwu

Hostility Detection and Covid-19 Fake News Detection in Social Media [article]

Ayush Gupta, Rohan Sukumaran, Kevin John, Sundeep Teki
2021 arXiv   pre-print
Using NLP techniques, we build a model that makes use of an abusive language detector coupled with features extracted via Hindi BERT and Hindi FastText models and metadata.  ...  Our model achieves a 0.97 F1 score on coarse grain evaluation on Hostility detection task. Additionally, we built models to identify fake news related to Covid-19 in English tweets.  ...  connected neural network for ensembling.  ... 
arXiv:2101.05953v1 fatcat:ttwfdvmgtvh5rm6uumvwlylf54

Tweetaneuse @ AMI EVALITA2018: Character-based Models for the Automatic Misogyny Identification Task (Short Paper)

Davide Buscaldi
2018 International Workshop on Evaluation of Natural Language and Speech Tools for Italian  
Our participation was focused on the use of language-independent, character-based methods. Italiano.  ...  Acknowledgments We would like to thank the program "Investissements d'Avenir" overseen by the French National Research Agency, ANR-10-LABX-0083 (Labex EFL) for the support given to this work.  ...  ., 2018) earlier this year with language-independent character-based models, based both on neural networks and classic machine learning algorithms.  ... 
dblp:conf/evalita/Buscaldi18 fatcat:ccewhv6izrcdvdpvmmrv5tl634

Content Noise Detection Model Using Deep Learning in Web Forums

Jiyoung Woo, Jaeseok Yun
2020 Sustainability  
For the deep learning model, raw text including and excluding special characters was utilized.  ...  Here, in this work, an automatic detection model for spam posts in web forums using both conventional machine learning and deep learning is proposed.  ...  The spam detection model, for example, cannot detect abusive posts or meaningless posts. This work aimed to develop a general model to be applied for multiple types of content noise.  ... 
doi:10.3390/su12125074 fatcat:tnzi6qarlncjpcevtwuup54ivm

Abusive Language Detection with Graph Convolutional Networks [article]

Pushkar Mishra, Marco Del Tredici, Helen Yannakoudakis, Ekaterina Shutova
2019 arXiv   pre-print
Previous research on automated abusive language detection in Twitter has shown that community-based profiling of users is a promising technique for this task.  ...  We show that such a heterogeneous graph-structured modeling of communities significantly advances the current state of the art in abusive language detection.  ...  Acknowledgments We would like to thank the anonymous reviewers for their useful feedback. Helen Yannakoudakis was supported by Cambridge Assessment, University of Cambridge.  ... 
arXiv:1904.04073v1 fatcat:6yhy2iajgvfwpkmjmo5t4r6m4q
« Previous Showing results 1 — 15 out of 3,603 results