This paper presents an in-depth study of ChatGPT, a state-of-the-art language model that is revolutionizing generative text. We provide a comprehensive analysis of its architecture, training data, and evaluation metrics and explore its advancements and enhancements over time. Additionally, we examine the capabilities and limitations of ChatGPT in natural language processing (NLP) tasks, including language translation, text summarization, and dialogue generation. Furthermore, we compare ChatGPT to other language generation models and discuss its applicability in various tasks. Our study also addresses the ethical and privacy considerations associated with ChatGPT and provides insights into mitigation strategies. Moreover, we investigate the role of ChatGPT in cyberattacks, highlighting potential security risks. Lastly, we showcase the diverse applications of ChatGPT in different industries and evaluate its performance across languages and domains. This paper offers a comprehensive exploration of ChatGPT’s impact on the NLP field.

In recent years, big data have become a hot research topic. The increasing amount of big data also increases the chance of breaching the privacy of individuals. Since big data require high computational power and large storage, distributed systems are used. As multiple parties are involved in these systems, the risk of privacy violation is increased. There have been a number of privacy-preserving mechanisms developed for privacy protection at different stages (e.g., data generation, data storage, and data processing) of a big data life cycle. The goal of this paper is to provide a comprehensive overview of the privacy preservation mechanisms in big data and present the challenges for existing mechanisms. In particular, in this paper, we illustrate the infrastructure of big data and the state-of-the-art privacy-preserving mechanisms in each stage of the big data life cycle. Furthermore, we discuss the challenges and future research directions related to privacy preservation in big data.

Insider threat detection has become a paramount concern in modern times where organizations strive to safeguard their sensitive information and critical assets from malicious actions by individuals with privileged access. This survey paper provides a comprehensive overview of insider threat detection, highlighting its significance in the current landscape of cybersecurity. The review encompasses a broad spectrum of methodologies and techniques, with a particular focus on classical machine-learning approaches and their limitations in effectively addressing the intricacies of insider threats. Furthermore, the survey explores the utilization of modern deep learning and natural language processing (NLP) based methods as promising alternatives, shedding light on their advantages over traditional methods. This analysis underscores the need for sophisticated solutions that can adapt to evolving threat landscapes and accommodate the intricacies of human behavior. In the conclusion section, the paper offers valuable insights into the future directions of insider threat detection. It advocates for the integration of more sophisticated time-series-based techniques, recognizing the importance of temporal patterns in insider threat behaviors. Additionally, the survey underscores the potential of NLP and large language model-based approaches, which can enhance threat detection by deciphering textual and contextual information. These recommendations reflect the evolving nature of insider threats and emphasize the need for proactive, data-driven strategies to safeguard organizations against internal security breaches. In conclusion, this survey not only underscores the urgency of addressing insider threats but also provides a roadmap for the adoption of advanced methodologies to enhance detection and mitigation capabilities in contemporary cybersecurity paradigms.

Due to the tremendous advancement in interactive multimedia systems and technologies, security has become a major aspect. Advanced technology can be utilized for hacking autonomous systems like Unmanned Aerial vehicles (UAVs) in different ways such as spoofing and jamming. It can be spoofed by the injection of fake signals into the sensors. For the protection of the UAVs from the Global Positioning System (GPS) signal spoofing attack, we propose a new methodology by incorporating a machine learning (ML) algorithm such as Support Vector Machine (SVM). A detailed analysis of several learning algorithms is also carried out to choose the suitable learning algorithm for the proposed work. Once the suitable ML algorithm is selected, we perform K-fold analyses to develop other learning models by choosing different values of K-folds thus we called them K-learning models. The purpose of developing K-learning models is to apply voting techniques to the developed K-learning models. Moreover, the signal features used in the proposed work are jitter, jitter (absolute), jitter (local), jitter (RAP), jitter (ppq5), shimmer, shimmer (local), shimmer (dB), shimmer (apq3), shimmer (apq5) and frequency modulation. Based on these features of the signal, we train our proposed model for the detection of counterfeit GPS signals. To gauge the performance of the proposed model, we perform different experimentation analyses such as accuracy, precision, recall, and F1-score. The results and analysis show the effectiveness of the proposed work over existing techniques.

With the rapid growth in technology, the use of Unmanned Aerial Vehicles (UAVs) have increased in civil and military applications including rescue operations, disaster recovery, and military operations. Despite the utility and advantages of UAVs, they may lead to major security breaches in the context of hardware, software, and communication channel, due their ease of use and availability. UAVs are vulnerable to various types of attacks such as spoofing, false data injection, jamming, fuzzing, availability, confidentiality, and integrity attacks. To overcome these security threats, researchers have been investigating strong security protocols to keep UAVs safe from the attackers. Nevertheless, there are many flaws in the developed protocols which can be exploited by hackers. Therefore, it is becomes crucial to study and analyze the existing security protocols used in UAVs to discover and address their vulnerabilities and weaknesses. The purpose of this study is to explore the vulnerabilities in the security protocols and propose guidelines to improve the security and provide future research directions.

In the contemporary landscape, where a huge amount of data plays a vital role, the importance of strong and robust cybersecurity measures has become increasingly paramount. This research proposes a review and extensively explores cybersecurity techniques within the domain of machine/deep learning and quantum techniques, with a particular focus on cryptographic methods and methodologies applied to image encryption. The proposed survey covers a range of cybersecurity techniques, including quantum random number generation, secure transmission of quantum images, watermarking through quantum methods, and quantum steganography. Moreover, it explores the domain of image encryption, which integrates adversarial neural networks, deep learning and machine learning, transformation techniques, and chaotic neural networks that can be used to secure digital data from cyber attacks. Our focus extends beyond highlighting advances in investigating vulnerabilities in existing cryptographic techniques. By identifying the challenges and weaknesses, the potential solutions are also presented, establishing a foundation for future recommendations. These future suggestions address and overcome the vulnerabilities observed in existing cybersecurity techniques. The aim of the extensive survey and analysis of existing cryptographic techniques is to provide a deep understanding of innovative and diverse approaches within the cybersecurity domain. Simultaneously, it aims to create a roadmap for the future to counter potential cyber threats and challenges.