Journal of Cyber Security, Privacy Issues and Challenges

Homomorphic Encryption for Edge Computing Security

S. Mohanraj — 2024-08-30

Homomorphic encryption is a cutting edge security method that can protect sensitive data by performing calculations on encrypted data directly without requiring decryption. This implies that businesses can uphold high data security without sacrificing productivity or adhering to legal requirements. Homomorphic encryption makes it possible to work with encrypted data while maintaining its confidentiality.
This study examines several Homomorphic Encryption (HE) techniques used in edge computing, including BGV, CKKS, RSA, Paillier, and Gentry's. These plans all have different benefits in terms of performance and security. While completely homomorphic encryption was first introduced by Gentry's method, which permits addition and multiplication operations on encrypted data, RSA and Paillier are recognized for their resilience. BGV and CKKS further enhance efficiency and practicality for real world applications.
Significant advantages of homomorphic encryption include decreased latency and enhanced privacy. Nevertheless, the study also discusses excessive energy usage and considerable storage needs. Applications of homomorphic encryption are practical and are emphasized in critical domains such as healthcare, where patient privacy is crucial; smart homes, where secure automation is needed; industrial IoT, where sensitive operational data needs to be protected; autonomous vehicles, where navigational data needs to be protected; and smart cities, where secure data is necessary for effective urban management.

Predictive Analytics through AI: Identifying Social Media Threat Patterns

Mohd. Asif Gandhi — 2024-11-19

The proliferation of social media platforms has introduced new challenges in monitoring and identifying potential threats, ranging from cyber-attacks to misinformation campaigns. This study explores using Artificial Intelligence (AI) for predictive analytics in identifying patterns associated with social media threats. Leveraging machine learning algorithms and Natural Language Processing (NLP), we developed an AI model capable of analyzing massive volumes of social media data to detect threat patterns and anticipate potential security risks. Our approach integrates sentiment analysis, keyword extraction, and network mapping to build predictive models that capture the complex dynamics of social media interactions. The proposed system was evaluated on a dataset collected from major social platforms over six months, focusing on detecting coordinated threats and abnormal user behaviors. Experimental results demonstrate that our AI model achieved an accuracy of 92% in identifying precursors to security threats, outperforming baseline methods by approximately 15%.
Furthermore, the model provides a scalable solution for real-time monitoring, making it feasible for deployment in various applications, from law enforcement to corporate security. The findings indicate that AI-driven predictive analytics can be a valuable tool for proactively addressing threats on social media, contributing to safer digital environments. Future research will enhance model robustness across different languages and social media platforms, improving cross-platform threat detection.

The Ripple Effect: An Analysis of Malware Spread in Connected IoT and Cyber-Physical Systems

P. Devi Sravanthi — 2024-11-28

The proliferation of the Internet of Things (IoT) and Cyber-Physical Systems (CPS) has introduced new complexities in cyber security, where interconnected devices can serve as vectors for malware propagation. This paper analyzes the dynamics of malware spread across IoT and CPS environments, emphasizing the ripple effect that compromises in one device can have on an entire network. By modeling the interdependencies among connected devices and systems, the study explores how vulnerabilities in seemingly isolated components can propagate, creating cascading failures across broader infrastructures. The analysis considers direct and indirect attack vectors, including system configurations, weak authentication protocols, and unsecured communication channels. Additionally, the paper examines how malicious software can exploit the unique characteristics of IOT/CPS networks such as real-time data processing, heterogeneous device types, and low-power constraints making traditional defense mechanisms inadequate. Using simulation-based experiments and case studies, we identify key factors that exacerbate the speed and scale of malware spread. Finally, the paper proposes mitigation strategies, including robust device authentication, anomaly detection systems, and network segmentation, to curb the impact of cyber attacks.

Cyber Sentinel: Monitoring and Rating Online Behavior to Combat Cyber Bullying

Suhas Amarnath — 2024-12-17

The proliferation of hate speech on social media platforms such as YouTube, Facebook, and Twitter pose a significant threat to societal harmony and disproportionately impacts marginalized communities. To address this problem, we provide a robust machine learning model that combines Convolutional Neural Networks (CNN) and Natural Language Processing (NLP) to identify bullying behavior on Twitter. Textual analysis has been the mainstay of traditional methods for identifying bullying on social media. While this approach has its benefits, it fails to consider the complex nature of online communication, which frequently involves visuals. Our approach integrates NLP to analyze the textual content of tweets, identifying potentially harmful language patterns and phrases indicative of bullying. Simultaneously, CNNs are employed to analyze images associated with tweets, identifying visual elements that may contribute to bullying. The NLP component of our model is designed to process and understand the nuances of natural language in tweets. This involves tokenization, sentiment analysis, and semantic similarity measures to classify tweets as bullying or non-bullying accurately. The CNN component learns to identify visual cues that may be signs of bullying behavior using a dataset of photos connected to bullying. Using the Twitter API, we continuously fetch tweets in real-time, ensuring our model remains up-to-date with the latest language and visual trends. The combined use of NLP and CNN enhances the model's accuracy in detecting true positives, thus providing a comprehensive solution to identify and mitigate social media bullying. Our project demonstrates the effectiveness of integrating textual and visual data analysis to detect bullying on social media platforms more accurately. Implementing this model could make social media environments safer and more inclusive, protecting users from the harmful impacts of online bullying.

A Comprehensive Study on Security in Edge-as-a-Service

R. Satya Pramodh — 2024-12-19

Edge-as-a-Service (EaaS) is a rapidly evolving framework that enables businesses to use edge computing capabilities without requiring extensive infrastructure investment. This method lowers latency and improves business performance by processing data closer to the source. EaaS's distributed architecture, however, creates serious security issues, such as flaws in network security, edge device integrity, and data transmission. To improve security in EaaS systems, this study examines many of the latest technologies. Key focus areas include secure data transmission, identity management, privacy preservation, and the integration of AI-driven security mechanisms. The paper also discusses future directions in enhancing the security of EaaS platforms.