Deep Multi-Task Models for Misogyny Identification and Categorization on Arabic Social Media (original) (raw)
Related papers
Let-Mi: An Arabic Levantine Twitter Dataset for Misogynistic Language
ArXiv, 2021
Online misogyny has become an increasing worry for Arab women who experience gender-based online abuse on a daily basis. Misogyny automatic detection systems can assist in the prohibition of anti-women Arabic toxic content. Developing such systems is hindered by the lack of the Arabic misogyny benchmark datasets. In this paper, we introduce an Arabic Levantine Twitter dataset for Misogynistic language (LeT-Mi) to be the first benchmark dataset for Arabic misogyny. We further provide a detailed review of the dataset creation and annotation phases. The consistency of the annotations for the proposed dataset was emphasized through inter-rater agreement evaluation measures. Moreover, Let-Mi was used as an evaluation dataset through binary/multi-/target classification tasks conducted by several state-of-the-art machine learning systems along with Multi-Task Learning (MTL) configuration. The obtained results indicated that the performances achieved by the used systems are consistent with ...
Tehnički glasnik
In recent years, social media networks are emerging as a key player by providing platforms for opinions expression, communication, and content distribution. However, users often take advantage of perceived anonymity on social media platforms to share offensive or hateful content. Thus, offensive language has grown as a significant issue with the increase in online communication and the popularity of social media platforms. This problem has attracted significant attention for devising methods for detecting offensive content and preventing its spread on online social networks. Therefore, this paper aims to develop an effective Arabic offensive language detection model by employing deep learning and semantic and contextual features. This paper proposes a deep learning approach that utilizes the bidirectional long short-term memory (BiLSTM) model and domain-specific word embeddings extracted from an Arabic offensive dataset. The detection approach was evaluated on an Arabic dataset coll...
Misogynistic Tweet Detection: Modelling CNN with Small Datasets
Data Mining, 2019
Online abuse directed towards women on the social media platform such as Twitter has attracted considerable attention in recent years. An automated method to effectively identify misogynistic abuse could improve our understanding of the patterns, driving factors, and effectiveness of responses associated with abusive tweets over a sustained time period. However, training a neural network (NN) model with a small set of labelled data to detect misogynistic tweets is difficult. This is partly due to the complex nature of tweets which contain misogynistic content, and the vast number of parameters needed to be learned in a NN model. We have conducted a series of experiments to investigate how to train a NN model to detect misogynistic tweets effectively. In particular, we have customised and regularised a Convolutional Neural Network (CNN) architecture and shown that the word vectors pre-trained on a task-specific domain can be used to train a CNN model effectively when a small set of labelled data is available. A CNN model trained in this way yields an improved accuracy over the state-of-the-art models.
Multi-Task Learning using AraBert for Offensive Language Detection
2020
The use of social media platforms has become more prevalent, which has provided tremendous opportunities for people to connect but has also opened the door for misuse with the spread of hate speech and offensive language. This phenomenon has been driving more and more people to more extreme reactions and online aggression, sometimes causing physical harm to individuals or groups of people. There is a need to control and prevent such misuse of online social media through automatic detection of profane language. The shared task on Offensive Language Detection at the OSACT4 has aimed at achieving state of art profane language detection methods for Arabic social media. Our team “BERTologists” tackled this problem by leveraging state of the art pretrained Arabic language model, AraBERT, that we augment with the addition of Multi-task learning to enable our model to learn efficiently from little data. Our Multitask AraBERT approach achieved the second place in both subtasks A & B, which s...
Proceedings of the 16th International Workshop on Semantic Evaluation (SemEval-2022)
This paper provides a comparison of different deep learning methods for identifying misogynous memes for SemEval-2022 Task 5: Multimedia Automatic Misogyny Identification. In this task, we experiment with architectures in the identification of misogynous content in memes by making use of text and image-based information. The different deep learning methods compared in this paper are: (i) unimodal image or text models (ii) fusion of unimodal models (iii) multimodal transformers models and (iv) transformers further pretrained on a multimodal task. From our experiments, we found pretrained multimodal transformer architectures to strongly outperform the models involving fusion of representation from both the modalities.
Leveraging Multi-domain, Heterogeneous Data using Deep Multitask Learning for Hate Speech Detection
2020
With the exponential rise in user-generated web content on social media, the proliferation of abusive languages towards an individual or a group across the different sections of the internet is also rapidly increasing. It is very challenging for human moderators to identify the offensive contents and filter those out. Deep neural networks have shown promise with reasonable accuracy for hate speech detection and allied applications. However, the classifiers are heavily dependent on the size and quality of the training data. Such a high-quality large data set is not easy to obtain. Moreover, the existing data sets that have emerged in recent times are not created following the same annotation guidelines and are often concerned with different types and sub-types related to hate. To solve this data sparsity problem, and to obtain more global representative features, we propose a Convolution Neural Network (CNN) based multi-task learning models (MTLs)! to leverage information from multip...
2021
In this paper, we address the problem of automatic misogyny identification focusing on understanding the representation capabilities of widely adopted embeddings and addressing the problem of unintended bias. The proposed framework, grounded on Sentence Embeddings and Multi-Objective Bayesian Optimization, has been validated on an Italian dataset. We highlight capabilities and weaknesses related to the use of pre-trained language, as well as the contribution of Bayesian Optimization for mitigating the problem of biased predictions.
2020
The pervasiveness of social networks in recent years has revolutionized the way we communicate. The chance is now opened up for every person to freely and anonymously share his thoughts, opinions and ideas in a real-time manner. However, social media platforms are not always considered as a safe environment due to the increasing propagation of abusive messages that severely impact the community as a whole. The rapid detection of abusive messages remains a challenge for social platforms not only because of the harm it may cause to the users but also because of its impact on the quality of service they provide. Furthermore, the detection task proves to be more difficult when contents are generated in a specific language known by its complexity, richness and specificities like the Arabic language. The aim of this paper is to provide a comprehensive review of the existing approaches for detecting abusive messages from social media in the Arabic language. These approaches extend from the use of traditional machine learning to the incorporation of the latest deep learning architectures. Additionally, a background on abusive messages and Arabic language specificities will be presented. Finally, challenges are described for better analysis and identification of the future directions.
Association for Computational Linguistics, 2019
This paper describes the submissions of our team, HAD-Tübingen, for the SemEval 2019-Task 6: "OffensEval: Identifying and Categorizing Offensive Language in Social Me-dia". We participated in all the three sub-tasks: Sub-task A-"Offensive language identifica-tion", sub-task B-"Automatic categorization of offense types" and sub-task C-"Offense target identification". As a baseline model we used a Long short-term memory recurrent neu-ral network (LSTM) to identify and categorize offensive tweets. For all the tasks we experimented with external databases in a postpro-cessing step to enhance the results made by our model. The best macro-average F 1 scores obtained for the sub-tasks A, B and C are 0.73, 0.52, and 0.37, respectively.
Empirical Analysis of Multi-Task Learning for Reducing Model Bias in Toxic Comment Detection
2019
With the recent rise of toxicity in online conversations on social media platforms, using modern machine learning algorithms for toxic comment detection has become a central focus of many online applications. Researchers and companies have developed a variety of shallow and deep learning models to identify toxicity in online conversations, reviews, or comments with mixed successes. However, these existing approaches have learned to incorrectly associate non-toxic comments that have certain trigger-words (e.g. gay, lesbian, black, muslim) as a potential source of toxicity. In this paper, we evaluate dozens of state-of-the-art models with the specific focus of reducing model bias towards these commonly-attacked identity groups. We propose a multi-task learning model with an attention layer that jointly learns to predict the toxicity of a comment as well as the identities present in the comments in order to reduce this bias. We then compare our model to an array of shallow and deep-lea...