International Journal of Networking and Virtual Organisations

Jakotiya, Komal; Shirsath, Vishal; Inamadar, Sharanbasawa

doi: 10.1504/ijnvo.2025.153489pmid: N/A

Distributed denial of service (DDoS) attack detection is necessitated as the security of the network data in recent decades is highly demandable. Several researches exist with numerous advantages but still contain certain challenges. To deal with the limitations and to perform significant attack detection, cannibalistic lead with long short-term memory hybridised explainable artificial intelligence (CLMO-XAI-SDBM) is proposed in the research. Further, the incorporation of the CLMO algorithm enhances the efficiency of the detection model as it remains the combination of characteristics of the bio-inspired algorithms. The classifier model included in the proposed research provides the advantages of handling the high-dimensional data and learning the multi-scaled features at different time series. The experimental results demonstrated that the proposed model attains high efficiency, which is evaluated with metrics such as precision, recall, and F1-score attaining 95.72%, 95.78%, and 95.71% respectively.

Begum, Nagma; Ahmad, Syed Jalal; Ishrathunissa,

doi: 10.1504/ijnvo.2025.153488pmid: N/A

The current surge in emerging technologies in networking has significantly improved people's lives by providing convenience and enjoyment. However, these technologies pose higher demands on network data processing and require a balance between security and stability. To meet the current load balancing needs, traditional network architectures use the link modules and overlook the server modules to improve the efficiency of the network. This study proposes the path-server traffic scheduling (PSTS) algorithm. It introduces server modules and utilises the software-defined network (SDN) paradigm with modular functionality implementation through the RYU controller. The procedure includes assessing performance metrics, specifically, impact factors at link and server levels. The algorithm ranks and filters each link and server by calculating weights based on previously obtained impact factor information, providing support for optimal traffic scheduling. Simulation results attain superior average bandwidth utilisation and reduced average transmission latency compared to the existing algorithm.

Yang, Liu

doi: 10.1504/ijnvo.2025.153496pmid: N/A

Although literature has shown that virtual communities (VC) and social media raise illness management awareness by enabling knowledge sharing, emotional support, and self-management tips. Drawing on social identity theory, this study specifically unfolds the associations among virtual communities, social media, social support, and chronic illness management (CIM) in the Chinese market. First, social media awareness was positively associated with CIM. Second, virtual communities and social support positively moderate the relationship between social media awareness and CIM. Besides, this study reports numerous applications for the management of CIM by emphasising virtual communities. In addition, this study has certain drawbacks that could be future research opportunities.

Aleem, Md; Ahmad, Sharjeel; Ahmad, Syed Jalal

doi: 10.1504/ijnvo.2025.153494pmid: N/A

Combining multiple learning models is an effective strategy for improving performance in complex decision-making systems. In belief propagation (BP) decoding, methods such as belief propagation list (BPL) and weighted belief propagation (WBP) provide complementary advantages in accuracy and adaptability. This work presents a novel decoding framework, termed ensemble weighted belief propagation (E-WBP), which integrates ensemble learning principles into the BP paradigm. The proposed method constructs multiple WBP decoders trained with diverse weight configurations during an offline phase. These decoders are then combined into a unified architecture that enhances decoding robustness. Performance results show that, for a fixed number of iterations, the proposed approach outperforms conventional BP decoding and approaches neural network-assisted WBP schemes, achieving an approximate gain of 0.3 dB over BPL at a frame error rate of 2 × 10-3. Furthermore, offline training reduces computational complexity and storage overhead, enabling efficient deployment in practical communication systems.