Paper Contents
Abstract
The widespread dissemination of extremist propaganda on social media platforms, namely Twitter, has presented substantial obstacles to upholding online safety and security. This review paper investigates the utilisation of deep learning methodologies to combine textual data and emotional indicators for the purpose of identifying extremist associates on the social media platform Twitter. Advancements in natural language processing (NLP) and affective computing have allowed for more advanced analyses of user-generated information. These studies may now capture not just the semantic meaning of the content, but also the underlying emotions being expressed. This method utilises neural network structures, specifically Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs), in combination with emotion recognition models to improve the precision of identifying extremist activity. The technique of combining textual and emotional analysis commences with pre-processing tweets to eliminate noise and extraneous information. Subsequently, deep learning models are utilised to extract features from both the textual content and the emotional range. The textual features encompass both syntactic and semantic aspectsTo summarise, the incorporation of text and emotions using deep learning algorithms provides a strong approach for detecting extremist associates on Twitter. The studies that were evaluated emphasise the potential of hybrid models in improving the ability to detect online extremism. This eventually helps social media companies and security authorities in their attempts to battle this issue. Subsequent investigations should prioritise the improvement of these models and tackling the ethical consequences of automated extremist detection in order to maintain a harmonious equilibrium between security and privacy.
Copyright
Copyright © 2024 PRABHJOT SINGH. This is an open access article distributed under the Creative Commons Attribution License.
