Multimodal cyberbullying detection using capsule network with dynamic routing and deep convolutional neural network

Abstract

Cyberbullying is the use of information technology networks by individuals' to humiliate, tease, embarrass, taunt, defame and disparage a target without any face-to-face contact. Social media is the "virtual playground" used by bullies with the upsurge of social networking sites such as Facebook, Instagram, YouTube, Twitter etc. It is critical to implement models and systems for automatic detection and resolution of bullying content available online as the ramifications can lead to a societal epidemic. This paper presents a deep neural model for cyberbullying detection in three different modalities of social data, namely, textual, visual and info-graphic (text embedded along with an image). The all-in-one architecture, CapsNet-ConvNet, consists of a Capsule network (CapsNet) deep neural network with dynamic routing for predicting the textual bullying content and a convolution neural network (ConvNet) for predicting the visual bullying content. The info-graphic content is discretized by separating text from the image using Google Lens of Google Photos App. The perceptron based decision-level late fusion strategy for multimodal learning is used to dynamically combine the predictions of discrete modalities and output the final category as bullying or non-bullying type. Experimental evaluation is done on a mix-modal dataset which contains 10000 comments and posts scrapped from YouTube, Instagram and Twitter. The proposed model achieves a superlative performance with the AUC-ROC of 0.98.

RETRACTED ARTICLE: CanarDeep: a hybrid deep neural model with mixed fusion for rumour detection in social data streams

“…The new feature vector has higher discriminative power in comparison to the individual input feature vectors. Whereas, in late fusion, individual classifiers are modelled using discrete input types and the predictions of the individual classifiers are combined to decide the final output [ 28 , 29 ]. Model-level fusion pools the benefits of both of these strategies by concatenating high-level feature representations from diverse classifiers.…”

Artificial Intelligence to Address Cyberbullying, Harassment and Abuse: New Directions in the Midst of Complexity

“…Furthermore, abusive words such as "bitch" or expletives can be used in a playful or friendly manner, resulting in false positives. To add to that, abusive content can be multimodal, involving text, image and video, where only one component can be abusive or none of the components is abusive on their own but rather they are only abusive when considered as a whole (Kumar & Sachdeva, 2021). For example, consider a photo of a tombstone with a comment underneath "you belong here" (Facebook AI, 2020).…”

A New Model for a Secure Social Media Application

“…It is obvious that a few characteristics (for example, the sender's followers) are powerfully connected with online bullying events according to the tested outcomes. The researchers in [13] proposed a DNN method for detecting cyberbullying in three distinct types of social information, such as visual media, infographics, and textual content (embedded text and an image). The compact structure, CapsNet-ConvNet, comprises ConvNet to predict the visual bullying content and CapsNet-DNN with dynamic routing to predict text bullying content.…”