A Review of Machine Learning Techniques in Cyberbullying Detection

Automatic identification of cyberbullying is a problem that is gaining traction,especially in the Machine Learning areas.Not only is it complicated,but it has also become a pressing necessity,considering how social media has become an integral part of adolescents’lives and how serious the impacts of cyberbullying and online harassment can be,particularly among teenagers.This paper contains a systematic literature review of modern strategies,machine learning methods,and technical means for detecting cyberbullying and the aggressive command of an individual in the information space of the Internet.We undertake an in-depth review of 13 papers from four scientific databases.The article provides an overview of scientific literature to analyze the problem of cyberbullying detection from the point of view of machine learning and natural language processing.In this review,we consider a cyberbullying detection framework on social media platforms,which includes data collection,data processing,feature selection,feature extraction,and the application ofmachine learning to classify whether texts contain cyberbullying or not.This article seeks to guide future research on this topic toward a more consistent perspective with the phenomenon’s description and depiction,allowing future solutions to be more practical and effective.

One Dimensional Conv-BiLSTM Network with Attention Mechanism for IoT Intrusion Detection

In the face of escalating intricacy and heterogeneity within Internet of Things(IoT)network landscapes,the imperative for adept intrusion detection techniques has never been more pressing.This paper delineates a pioneering deep learning-based intrusion detection model:the One Dimensional Convolutional Neural Networks(1D-CNN)and Bidirectional Long Short-Term Memory(BiLSTM)Network(Conv-BiLSTM)augmented with an Attention Mechanism.The primary objective of this research is to engineer a sophisticated model proficient in discerning the nuanced patterns and temporal dependencies quintessential to IoT network traffic data,thereby facilitating the precise categorization of a myriad of intrusion types.Methodology:The proposed model amal-gamates the potent attributes of 1D convolutional neural networks,bidirectional long short-term memory layers,and attention mechanisms to bolster the efficacy and resilience of IoT intrusion detection systems.A rigorous assessment was executed employing an expansive dataset that mirrors the convolutions and multifariousness characteristic of genuine IoT network settings,encompassing various network traffic paradigms and intrusion archetypes.Findings:The empirical evidence underscores the paramountcy of the One Dimensional Conv-BiLSTM Network with Attention Mechanism,which exhibits a marked superiority over conventional machine learning modalities.Notably,the model registers an exemplary AUC-ROC metric of 0.995,underscoring its precision in typifying a spectrum of intrusions within IoT infrastructures.Conclusion:The presented One Dimensional Conv-BiLSTM Network armed with an Attention Mechanism stands out as a robust and trustworthy vanguard against IoT network breaches.Its prowess in discerning intricate traffic patterns and inherent temporal dependencies transcends that of traditional machine learning frameworks.The commendable diagnostic accuracy manifested in this study advocates for its tangible deployment.This investigation indubitably advances the cybersecurity domain,amplifying the fortification and robustness of IoT frameworks and heralding a new era of bolstered security across pivotal sectors such as residential,medical,and transit systems.