Asian Journal of Research in Computer Science
Not a member yet
792 research outputs found
Sort by
Advanced Machine Learning for Robust Botnet Attack Detection in Evolving Threat Landscapes
No full textAs technology advances and security issues and cyberattacks increase, extensively Internet of Things (IoT) devices are linked to networks, and botnets have been emerging and evolving very fast, and they pose a dangerous threat. As systems become more complex, scale and, therefore, more complex, cyberattacks mounted against their vulnerabilities also increase. IoT transition is disrupted using these attacks, disrupting the IoT devices\u27 networks and services approaches for botnet attack detection and classification using Machine Learning (ML) and Deep Learning (DL) have been developed within the framework of the IoT. This study provides an intrusion detection system (IDS) based on the Bidirectional Gated Recurrent Unit (Bi-GRU) for detecting botnet attacks in IoT networks. We use the N-BaIoT dataset for this purpose. The study opted for a Bi-GRU model, which can detect contextual dependencies in the past and the future, to deal with the sequential IoT traffic data. The Bi-GRU model performance achieved exceptional results in classifying network traffic. The system\u27s accuracy in identifying both malicious and benign traffic was 99.99%. Additionally, the accuracy of these models rapidly rises and eventually levels out at almost 100%, indicating strong model performance. The model\u27s ability to recognise various botnet attack types even in cases of data imbalance was demonstrated by important performance metrics such as ROC-AUC, accuracy, precision, recall, and F1-score. The results show that the proposed Bi-GRU-based IDS is a robust and improved solution for detecting IoT botnet attacks on a real-time basis. While the model performs impressively, it has some problems, including the minor misclassification in complex attack cases and dependency on a single dataset, which restricts its generalisation. Future work will focus on improving model robustness
The Dead Internet Theory: A Survey on Artificial Interactions and the Future of Social Media
No full textThe Dead Internet Theory (DIT) suggests that much of today’s internet, particularly social media, is dominated by non-human activity, AI-generated content, and corporate agendas, leading to a decline in authentic human interaction. This study explores the origins, core claims, and implications of DIT, emphasizing its relevance in the context of social media platforms. The theory emerged as a response to the perceived homogenization of online spaces, highlighting issues like the proliferation of bots, algorithmically generated content, and the prioritization of engagement metrics over genuine user interaction. AI technologies play a central role in this phenomenon, as social media platforms increasingly use algorithms and machine learning to curate content, drive engagement, and maximize advertising revenue. While these tools enhance scalability and personalization, they also prioritize virality and consumption over authentic communication, contributing to the erosion of trust, the loss of content diversity, and a dehumanized internet experience. This study redefines DIT in the context of social media, proposing that the commodification of content consumption for revenue has taken precedence over meaningful human connectivity. By focusing on engagement metrics, platforms foster a sense of artificiality and disconnection, underscoring the need for human-centric approaches to revive authentic online interaction and community building
AI-Driven Open Source Intelligence in Cyber Defense: A Double-edged Sword for National Security
No full textThis study explores the dual implications of Artificial Intelligence (AI)-driven Open Source Intelligence (OSINT) in enhancing cyber defense capabilities. Using publicly available datasets, including IBM X-Force breach metrics, MITRE ATT&CK adversarial tactics, GDPR privacy violations, AI-driven phishing incidents, and case-specific data from the Colonial Pipeline ransomware attack and Russia-Ukraine conflict, the research employs multivariate regression, logistic regression, and K-Means clustering. The findings indicate that AI investments improve detection time (-0.68), accuracy (+2.09), and resolution rates (+1.55) with statistical significance (p < 0.001). However, risks associated with algorithmic opacity, weak regulatory frameworks, and reactive AI systems pose ethical and operational challenges. Clustering reveals variability in AI applications, with optimized systems achieving 95.2% detection rates and 5.5-hour response times. Recommendations include investing in scalable tools, strengthening regulations, fostering public-private collaborations, and enhancing reactive AI oversight. The results highlight AI’s transformative potential in cyber defense while emphasizing the need for ethical and regulatory alignment. Future directions include testing these models in diverse operational environments to validate effectiveness and exploring hybrid AI approaches to balance proactive and reactive capabilities, ensuring robust and adaptive defense mechanisms
Comparing Traversal Strategies: Depth-first Search vs. Breadth-first Search in Complex Networks
No full textThis article compares and contrasts two basic graph traversal algorithms that are commonly employed in computational problem-solving and network research. Common applications of these algorithms include pathfinding, optimisation of network flows, collaborative exploration, and classification tasks. To find out how well they function with different types of datasets, network topologies, and issue domains, researchers have systematically reviewed previous works. We measured the efficiency of each solution using performance indicators like execution time, memory utilisation, and path length. According to the results, one approach is more effective in memory-constrained settings and deep searches, while the other is better at discovering the shortest paths and providing comprehensive coverage. Furthermore, the paper emphasises the advantages of hybrid techniques, which merge the best features of both algorithms to provide better results in specific cases. This comparison helps fill gaps in our knowledge of graph-based problem-solving methods and sheds light on how to choose the best traversal algorithms for different types of applications
Advancing Cybersecurity through Machine Learning: Bridging Gaps, Overcoming Challenges, and Enhancing Protection
No full textThe greatest technical achievement of the twenty-first century is machine learning (ML). The application of machine learning to detect cybersecurity vulnerabilities is a significant advancement in information security. A void exists in the field since the widespread application of machine learning technologies in cybersecurity remains distant. The primary cause of this gap is that contemporary technology has rendered it challenging for people to comprehend the role of machine learning in cybersecurity.
The review seeks to furnish readers with a comprehensive analysis of machine learning\u27s relevance across several facets of information security, especially for individuals interested in cybersecurity. It highlights the benefits of machine learning compared to human-operated detection methods and the diverse cybersecurity tasks it can do. This research elucidates various fundamental issues that impact real-world machine learning applications in cybersecurity. Ultimately, it examines how diverse businesses might advance machine learning in cybersecurity in the future, as this is crucial for the field\u27s further growth. This study analyzes the contribution of machine learning to the enhancement of cybersecurity, highlighting the necessity of safeguarding sensitive information from theft and loss, as well as protecting critical assets against cyberattacks
Adaptive Search Algorithms: A Comprehensive Overview and Emerging Optimization Trends
No full textAdaptive search algorithms have emerged as pivotal tools for addressing complex, high-dimensional, and nonlinear challenges across various domains. This paper provides a detailed review of adaptive search techniques, including evolutionary algorithms, swarm intelligence methods, and cutting-edge hybrid models, with a unique contribution of a systematic comparison that showcases quantifiable improvements—up to 50% reduction in computational overhead and a 30% increase in solution accuracy across diverse benchmarks. It delves into key methodologies such as genetic algorithms, particle swarm optimization, and differential evolution, highlighting recent breakthroughs in adaptive parameter tuning and multi-objective optimization frameworks. The research emphasizes significant advancements in practical applications like machine learning, engineering design, and logistics, where these algorithms have improved the balance between exploration and exploitation for more optimal outcomes. Furthermore, emerging trends such as bio-inspired models and the integration of reinforcement learning and quantum-enhanced optimization are discussed, promising to reshape the adaptive search landscape by equipping it with sophisticated tools to manage the growing complexity of optimization challenges. This paper aims to map the current state and guide future directions of adaptive search algorithms, fostering the development of more robust, efficient, and adaptable optimization strategies essential for ongoing academic and practical innovations
A Review of Reinforcement Learning: Current Trends and Future Prospects in Autonomous Systems
No full textThis review focuses on the use of reinforcement learning (RL) for autonomous systems and current trends and future prospects. It is therefore the intended goal to critically evaluate the concept of RL for improving autonomous decision making with focus on current and emerging issues including; sample efficiency, scalability, and safety. This review methodology is a synthesis of 10 studies which has been conducted between the years 2021 and 2024. However, these are some of the challenges that seem to plague RL even as it has potential to be used in realistic applications such as robots, self-driving cars and smart grid. The review also opines that due to developments of algorithms, computer intrinsics and safety mechanism, RL perhaps holds the key to the future for autonomous systems
Automated Claims Processing in Guidewire ClaimCenter: Enhancing Efficiency and Accuracy in the Insurance Industry
No full textAims: This study explores the benefits, challenges, and future trends in implementation of automated claims processing in Guidewire ClaimCenter, a leading software platform for insurance providers and provides insights to help insurers who intend to implement automated claims processing in Guidewire ClaimCenter.
Study Design: Mixed-methods approach, combining both qualitative and quantitative research to provide a comprehensive analysis of automation in Guidewire’s claims processing.
Place and Duration of Study: Analysis between February 2024 and September 2024, based on data from North America, Europe, and Asia-Pacific insurance markets as documented in vendor case studies, expert interviews, customer testimonials, and industry reports.
Methodology: Reviewed Guidewire product documentation, industry reports, whitepapers, and research papers related to digital transformation in insurance. Case study analysis from insurance companies that have implemented Guidewire ClaimCenter for claims automation, like Nationwide, AXA, and Liberty Mutual. Key performance indicators analyzed include claims processing time, cost savings, fraud detection, and customer experience improvements. Quantitative analysis included an online survey targeting about 100 professionals like claim adjusters, underwriters, and claim managers in insurance companies using Guidewire ClaimCenter and the questions focused on customer satisfaction, efficiency improvements, and cost savings.
Results: Automated processing of claims in Guidewire ClaimCenter resulted up to 50% reduction in claim settlement time for standard claims, enhanced fraud detection, improved customer satisfaction, and reduced adjuster workload by up to 30%. Challenges with automated claims processing includes integration complexities, workforce adaptation, AI limitations, and handling complex claims.
Conclusion: Guidewire ClaimCenter’s automation capabilities are transforming claims processing by enhancing efficiency, reducing costs, and improving policyholder satisfaction. Case studies from insurers demonstrate how Guidewire’s automation has led to faster settlements, reduced fraud, and improved customer experiences. Insurers investing in Guidewire ClaimCenter’s automation capabilities will be well-positioned to stay competitive in the evolving digital landscape
Real-World Implementations of Network Centrality Algorithms across Various
No full textThe tutorial analyses how and in what ways the algorithms are used and can be employed in different fields and in the scientific context. This paper revisits the centrality measures; degree, betweenness, closeness and eigenvector centrality and how they are used to study complex networks from social networks, to biology, economy and telecommunications. From the couple of cases and examples, the article shows how these algorithms are utilized to find these vital nodes, improve network performance as well as decision making. Pros and Cons Each of the methods and results evidence the flexibility of centrality measures in enhancing the knowledge and analysis of complex systems. However, issues of scaling these algorithms to operational networks and environments are still explicit. The need for tailor-made centrality analysis methods of scalability and adaptiveness are also outlined for future research agendas in this review
Deep Learning Techniques for Threat Detection in Cloud Environments: A Review
No full textDeep learning techniques have become essential in enhancing threat detection within cloud environments, offering the ability to process large-scale data and detect complex patterns. As cloud computing continues to grow, ensuring robust security measures is critical to protecting sensitive data from evolving cyber threats. Deep learning models, particularly CNN, RNN, and Autoencoders, play a key role in identifying various threats, such as unauthorized access, data leakage, and DDoS attacks. This paper reviews research published between 2018 and 2023, comparing the effectiveness of deep learning models in cloud security. The findings indicate that deep learning models provide higher accuracy and adaptability compared to traditional methods. However, challenges such as data confidentiality, high computational requirements, and real-time detection still persist. The paper concludes by highlighting the need for hybrid models and enhanced training datasets to overcome these challenges. This review is valuable for researchers and practitioners working to implement deep learning approaches in cloud security