Enhanced Real-Time Detection of Cyber Threats through Adaptive Machine Learning in Network Traffic Analysis

C.; A. R. Darshika; Keirolona Safana

doi:https://doi.org/10.54216/JCHCI.100102

Fusion: Practice and Applications | 2026

Using Deep Learning Strategy to Implement AI Tools Fusion in Academics

The advancement of artificial intelligence (AI) in the field of education system has revolutionized traditional education paradigms. The ability of language models to process human language has revolutionized the field of artificial intelligence. The fusion of deep learning and cognitive science is getting attention in the academic system. The absence of structured policies and lack of AI fusion strategies in academics disrupt traditional teaching classrooms resulting in misuse and resistance in adoption of AI. This marks the importance of preparation of AI policies for effective implementation of AI tools in teaching and learning. This paper highlights the importance of framing the guidelines for organized and practical implementation of AI fusion in academics. This study bridges the gap by developing a standardized framework to transform normal classrooms into dynamic data driven platforms promoting professional development for teachers and empowering students with digital literacy and autonomous learning. The study examines predictive performance using deep learning strategies to extract key features of teaching, learning and cognitive and predicts the impact of AI in sustainable teaching.   The highest importance scores range from 0.89 to 0.94, which indicates the importance of selected key features in models’ predictions. The highest mean score of 4.5 of the model establishes satisfaction of teachers and students with policy objectives. The results of the study indicate that integration of deep learning cognitive strategy along with clear policies framework help in achieving higher adoption and performances rates of AI in sustainable classrooms when compared with traditional teaching strategies with minimal AI-integration.

groups

Moosa Ahmed Hassan Bait Ali Sulaiman mail -

Anita Venugopal mail

link https://doi.org/10.54216/FPA.210119

Volume & Issue

Vol. Volume 21 / Iss. Issue 1

Details open_in_new

Prospects for Applied Mathematics and Data Analysis | 2025

A Brief Study on Fuzzy Off-Group Theory

Uncertainty-handling frameworks such as fuzzy sets, rough sets, intuitionistic fuzzy sets, neutrosophic sets, Picture Fuzzy Sets, hyperneutrosophic sets, and plithogenic sets have attracted sustained research interest. These frameworks have been widely applied across various mathematical disciplines, including graph theory, topology, algebra, and group theory. More recently, the concept of the offset has emerged as a powerful and promising generalization of conventional uncertainty models. In this paper, we introduce a novel algebraic structure called the Fuzzy Off-Group and conduct an in-depth study of its fundamental mathematical properties. We hope that this framework will further advance research in group theory and uncertainty modeling with offsets, and that it will open up new avenues for application.

groups

Takaaki Fujita mail -

Arif Mehmood Khattak mail -

Arkan A. Ghaib mail

link https://doi.org/10.54216/PAMDA.050101

Volume & Issue

Vol. Volume 5 / Iss. Issue 1

Details open_in_new

Prospects for Applied Mathematics and Data Analysis | 2025

On the Properties and Illustrative Examples of Soft SuperHypergraphs and Rough SuperHypergraphs

In graph theory, a hypergraph generalizes a classical graph by allowing each hyperedge to join any number of vertices, thereby modeling relationships beyond simple pairwise connections.[1] A superhypergraph takes this further by applying recursive powerset constructions to its hyperedge set, creating hierarchical and self-referential network layers.[2] A soft graph defines a family of subgraphs parameterized over a fixed universe of vertices and edges, while a rough graph uses lower and upper approximations to capture uncertainty in graph structure. In this paper, we revisit Soft SuperHypergraphs and Rough SuperHypergraphs—originally introduced in [3]—which integrate the flexibility of soft and rough graph frameworks with the layered com- plexity of superhypergraphs. We provide precise definitions, illustrative examples, and a detailed analysis of their fundamental properties, demonstrating their potential for modeling hierarchical and uncertain network systems.

groups

Takaaki Fujita mail -

Atiqe Ur Rahman mail -

Arkan A. Ghaib mail -

Talal Ali Al-Hawary mail -

Arif Mehmood Khattak mail

link https://doi.org/10.54216/PAMDA.050102

Volume & Issue

Vol. Volume 5 / Iss. Issue 1

Details open_in_new

International Journal of Neutrosophic Science | 2026

The Role of Neutrosophic Logic in Enhancing Trust and Reliability in Internet of Things Architectures

A vast amount of Internet of Things (IoT) devices deployment has created huge issues about trust management and reliability guarantees in heterogeneous, dynamic and often uncertain ecosystems. Available probabilistic or fuzzy-logic-based models do not hold water to deal with indeterminacy and contending data in distributed IoT networks. The current paper proposes a brand new framework to model trust and reliability in IoT systems by implementing Neutrosophic Logic to build quantification and strengthen trust and reliability in IoT systems. Incorporating the semantic understanding of data and node behavior in uncertainty using three dissimilar elements to represent trust: truth, indeterminacy and falsity, the model commands a wider range of semantics in the relationship of data and nodes during the phase of uncertainty. A mathematical solution is established to measure trust scores and reliability indexes based on Neutrosophic membership functions. Further, a new dynamic trust assessment and anomaly detection algorithm is presented based on a multi-layered decision-making process. This simulation and case- study definition shows the effectiveness of the proposed framework in having less false positives, better reliability estimation, and the solid optimization of decision support in a very uncertain environment of IoT. The work therefore further develops the process of Neutrosophic systems integration with IoT and its setting up of basis of more intelligent, context-aware and robust trust management systems.

groups

Remya P. George mail -

Nazia Ahmad mail -

Rubina Liyakat Khan mail -

Sajithunisa Hussain mail -

Samandarboy Sulaymanov mail -

Ambuj Kumar Agarwal mail

link https://doi.org/10.54216/IJNS.270117

Volume & Issue

Vol. Volume 27 / Iss. Issue 1

Details open_in_new

Journal of Intelligent Systems and Internet of Things | 2026

Energy-Aware UAV Relaying with SWIPT and Real-Time Reinforcement Learning for Disaster Response

Wireless sensor networks used in disaster-struck areas experience the problem of energy constraints, which may negatively affect the data communication process. A novel energy-aware UAV relaying scheme is presented that incorporates SWIPT (Simultaneous Wireless Information and Power Transfer) to power the UAVs and their ground sensor devices. Dynamic power and flight path allocation according to the environmental conditions is achieved with dynamic reinforcement learning and, in particular, with a Proximal Policy Optimization (PPO) method. The system maximizes energy gathering at the sensor nodes and lengthens UAV flight life, and preserves high-quality signal transmission. The findings indicate a 23.5 dB increase in the SINR, 83.2 percent efficiency of energy harvesting, and an average of 43.2 minutes of endurance for the UAV. The success rate on the relay was 94.6 per cent, and a convergence of 12.3 seconds. The model also took the lead over other past ways in terms of mission coverage and energy efficiency in various simulation cases. This system enhances the resilience of disaster communication by effectively utilizing energy resources. Finally, it makes adaptation in real time and continued work in high-danger situations possible.

groups

P. Keerthana mail -

A. Vijayalakshmi mail

link https://doi.org/10.54216/JISIoT.180127

Volume & Issue

Vol. Volume 18 / Iss. Issue 1

Details open_in_new

Journal of Intelligent Systems and Internet of Things | 2026

A Deep Convolutional Autoencoder with Metaheuristic Optimization based Feature Reduction Framework for Genetic Disorder Detection Model

Genetic disorder is an outcome of transformation in deoxyribonucleic acid (DNA) system, which is progressed or natural from blood relation. Such transformations might lead to deadly illnesses like Alzheimer’s, cancer, and much more. The disorder of single gene kind is affected by a change in a solitary gene in DNA. The chromosomal disorder kind is affected when a genetic material or a portion of chromosome is removed or substituted in the structure of DNA. Complex illnesses are caused by the alteration in over one gene exhibit in the DNA. In recent times, the usage of artificial intelligence (AI)-based deep learning (DL) systems has exposed excellent achievement in the prognosis and prediction of diverse illnesses. The latent of DL models are employed to forecast genetic disorder at an initial phase utilizing the genome data for appropriate treatment. This paper presents a Deep Feature Selection Framework for Genetic Disorder Detection Using Convolutional Autoencoder and Metaheuristic Optimization (DFSFGDD-CAEMO) model. The aim of DFSFGDD-CAEMO model is to develop an accurate DNA-based genetic disorder classification model using advanced techniques for early and reliable disease diagnosis. Initially, the min-max normalization method is employed in the data pre-processing stage for converting an input data into a beneficial format. Besides, the Aquila optimizer (AO) method has been deployed for the selection of feature process in order to select the most significant features from a dataset. For the classification procedure, the proposed DFSFGDD-CAEMO technique designs Convolutional Autoencoder (CAE) method. At last, the hyperactive parameter tuning process is performed through enhanced pelican optimization algorithm (EPOA) for improving the classification performance of CAE model. The experimental evaluation of the DFSFGDD-CAEMO technique occurs using benchmark dataset. The experimentation results indicated out the enhanced performance of the DFSFGDD-CAEMO system when equated to existing approaches.

groups

S. Puvaneswari mail -

G. Indırani mail

link https://doi.org/10.54216/JISIoT.180128

Volume & Issue

Vol. Volume 18 / Iss. Issue 1

Details open_in_new

Fusion: Practice and Applications | 2026

An Advanced Immersion Level Prediction Using Ensemble Classifier with Heuristic Search Algorithm in 3D Games Content Generated Virtual Environments

Nowadays, virtual reality (VR) and immersive environments are research fields used in various educational and scientific areas. Immersive digital media desires new techniques for its immersive and interactive features it implies the model of new relationships and narratives with users. VR and technologies related to the virtuality sequence, like digital and immersive environments, are developing media. 3D environments generated with VR compatibility can be skilled from a stereoscopic and egocentric view that outperforms the immersion of the ‘classical’ screen-based view of 3D gamed virtual environments. Recent video games have complete, interactive scenes generated with innovative modeling and animation software and provided with hardware speeded-up graphics and physics. Their communication takes place with body-based sensing and commodity 3D motion controllers, like and in certain ways more progressive, than those discovered in conventional VEs do. Currently, artificial intelligence-based deep learning (DL) methods have been progressively applied to identify and assess user immersion levels in VR environments. In this paper, we present an Advanced Immersion Level Prediction Using Ensemble Classification Model and Metaheuristic Optimization Algorithm (ILPECM-MOA) in 3D Games Virtual Environments. This paper aims to develop a predictive model for assessing advanced immersion levels in 3D game virtual environments using behavioral and contextual data. At the primary stage, the data pre-processing stage uses Z-score normalization to transform input data into a beneficial pattern. Followed by, the presented ILPECM-MOA method designs ensemble models such as the temporal convolutional network (TCN) model, sparse denoising autoencoder (SDAE) method, and stacked long short-term memory (SLSTM) technique for the classification process. At last, the Hybrid ebola and Bald Eagle search optimization (HEBEO) approach fine-tunes the hyperparameter values of ensemble methods and results in the superior performance of classification. The effectiveness of the ILPECM-MOA model has been validated by the detailed studies utilizing the benchmark dataset. The mathematical outcome indicates that the ILPECM-MOA approach has improved performance and scalability in terms of various measures over the recent methods.

groups

Kamalanathan Sundararajan mail -

Prasanna Santhanam mail

link https://doi.org/10.54216/FPA.210120

Volume & Issue

Vol. Volume 21 / Iss. Issue 1

Details open_in_new

Fusion: Practice and Applications | 2026

A Model for the Prediction of Cardiovascular Disease in IoMT Based on AI's Binary and Multi-Class Structures

Heart disease is a severe hazard to the public's health and safety because of the high rates of disability and mortality it causes. Accurate disease prediction and diagnosis are more critical than ever in this era of earlier illness prevention, faster disease detection, and earlier disease treatment. Artificial Intelligence (AI) and the Internet of Medical Things (IoMT) have made it possible to detect, forecast, and diagnose cardiovascular disease more precisely. However, the bulk of these prediction models can only state whether a person is sick; they cannot and do not forecast the severity of the ailment. We present a machine-learning-based technique for predicting cardiovascular disease. Using this strategy, we hope to perform binary and multimodal classifications at the same time. To get things started, we will go through the fuzzy-adaboost approach, which will serve as the foundation for the rest of our work. By combining fuzzy logic and the Adaboost method, this method aims to increase the number of applications that can use binary classification prediction to simplify data analysis. If it is completed, both objectives will be met, and we will eliminate overfitting by merging bagging and fuzzy adaboost into a single approach. It is the ideal solution to the challenge we are currently facing. Because it has a separate classification for the severity of the presentation of heart disease, the bagging fuzzy adaboost can be used for multiclassification prediction. This is because Adaboost's assessment of the severity of the observed heart disease presentations is unclear and imprecise. The results of the experiment reveal that, in addition to a wide range of other classes, the Bagging-Fuzzy-Adaboost can anticipate binary data accurately. When compared to traditional procedures, it is evident that this has significant advantages.

groups

Ahmed A. F. Osman mail -

Nesren Farhah mail -

Rajit Nair mail -

Mohammed Awad Mohammed Ataelfadiel mail -

Rami Taha shehab mail

link https://doi.org/10.54216/FPA.210121

Volume & Issue

Vol. Volume 21 / Iss. Issue 1

Details open_in_new

Journal of Cognitive Human-Computer Interaction | 2025

HeartLink: IoT Smartwatch for Emergency Alerts

This paper introduces HeartLink, an IoT-based health monitoring system designed to provide real-time heart rate tracking and emergency alerts using the Huawei Band 9 smartwatch. The system integrates Huawei's Health Kit with a Flutter-based Android application, enabling seamless data collection and processing. The backend, built on Java Spring Boot or Node.js, utilizes a hybrid database architecture combining MongoDB and Firebase for efficient data storage and real-time synchronization. HeartLink features threshold- based alert mechanisms, where heart rate deviations trigger SMS notifications to pre-selected contacts via Twilio and emergency calls to ambulance services in critical scenarios. Firebase Cloud Messaging (FCM) ensures timely push notifications, while Firebase Authentication secures user access. The system's modular design allows for real-time heart rate analysis, dynamic threshold configuration, and automated emergency responses, making it a robust solution for individuals requiring continuous health monitoring. By leveraging advanced IoT and cloud technologies, HeartLink bridges the gap between wearable health devices and emergency response systems, offering a scalable, reliable, and user-friendly platform for real-time health tracking and life- saving interventions.

groups

A. V. Adlin Grace mail -

Cherlin Flory Thomas mail -

Anu Sushmitha S. mail

link https://doi.org/10.54216/JCHCI.100101

Volume & Issue

Vol. Volume 10 / Iss. Issue 1

Details open_in_new

Journal of Cognitive Human-Computer Interaction | 2025

Enhanced Real-Time Detection of Cyber Threats through Adaptive Machine Learning in Network Traffic Analysis

As cyber threats become more complex, real-time systems are needed to detect and eliminate attacks. Traditional network intrusion detection systems based on rule based static method tend to be ineffective against novel emerging threats. In this paper, we propose an improved real time cyber threat detection system using adaptive machine learning techniques used to analyze network traffic and find anomalies. Our proposed approach uses a blend of supervised and unsupervised learning models such that the system maintains high detection accuracy with minimal false positives, while maintaining continuous adaptation to constantly evolving threats. On critical network traffic features like packet size, flow duration, source and destination IP addresses, transmission protocols, the system is then trained. They show experimentally better detection accuracy, responsiveness and adaptability than conventional IDS. In this work, contributions of adaptive machine learning for robustness against dynamic and evolving threats in network environments are highlighted as significant strides towards improving real time cybersecurity infrastructure.

groups

C. Meenaloshini mail -

A. R. Darshika Kelin mail -

Keirolona Safana Seles mail

link https://doi.org/10.54216/JCHCI.100102

Volume & Issue

Vol. Volume 10 / Iss. Issue 1

Details open_in_new