International Journal of Cognitive Computing in Engineering最新文献

{"title":"High-performance image encryption framework using 2D nonuniform FFT and TDES for securing cognitive applications","authors":"Mohamed Ayari ,&nbsp;Yamen El Touati ,&nbsp;Atef Gharbi ,&nbsp;Zeineb Klai","doi":"10.1016/j.ijcce.2025.11.003","DOIUrl":"10.1016/j.ijcce.2025.11.003","url":null,"abstract":"<div><div>This study proposes a fast, hybrid image cryptosystem that pairs classical encryption with transform-domain processing for securing multimedia and cognitive-computing use cases. The pipeline combines the 2D nonuniform FFT (2D-NUFFT) for nonuniform spectral scrambling, 2D discrete wavelet transform (2D-DWT) for multiresolution decomposition, chaotic random phase mask (CRPM) for nonlinear diffusion, and triple DES (TDES) for block-cipher protection. The results of simulations conducted show lossless reconstruction (MSE ≈ <span><math><msup><mrow><mn>10</mn></mrow><mrow><mo>−</mo><mn>24</mn></mrow></msup></math></span>; PSNR = ∞, SSIM = 1.000) and strong security: NPCR = 99.65 %, UACI = 33.52 %, entropy = 7.9963, near-zero correlation (0.0009), and a near-ideal avalanche (50.06 %). The overall computational cost is <span><math><mrow><mi>O</mi><mo>(</mo><mrow><msup><mrow><mi>N</mi></mrow><mn>2</mn></msup><mtext>log</mtext><mi>⁡</mi><mi>N</mi></mrow><mo>)</mo></mrow></math></span> (with the NUFFT as the main contributor), and a 512 × 512 frame completes in approximately 172 ms on a CPU-only setup. Taken together, these results indicate that the proposed framework provides a reliable, lightweight solution for real-time, high-security applications while maintaining exact decryption and robust resistance to common cryptanalytic attacks.</div></div>","PeriodicalId":100694,"journal":{"name":"International Journal of Cognitive Computing in Engineering","volume":"7 ","pages":"Pages 287-299"},"PeriodicalIF":0.0,"publicationDate":"2026-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145623221","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Optimization and performance analysis of Drones and Unmanned Aerial Systems and Their Intelligence Applications","authors":"Ibrahim Abaker Hashem ,&nbsp;Saber Zerdoumi ,&nbsp;NZ Jhanjhi ,&nbsp;Aisha Siddiqa ,&nbsp;Sebti Foufou","doi":"10.1016/j.ijcce.2025.10.004","DOIUrl":"10.1016/j.ijcce.2025.10.004","url":null,"abstract":"<div><div>Advancements in unmanned aerial systems (UAS) and artificial intelligence (AI) have emerged in recent years, which have accelerated research in a variety of fields, including human-drone interaction, autonomous navigation, security, object detection, urban air mobility, energy-efficient design, environmental monitoring, archaeological research, wildlife conservation, medical supply delivery, disaster response, and precision agriculture. These represent just a few areas where AI-driven drones are making an impact in society. The study discussed the main principles of autonomous navigation and its mathematical aspects and presented the state-of-the-art algorithms in unmanned aerial vehicles (UAVs). The study also demonstrated an optimization model for path loss considering terrain and environmental constraints and provided a comparative analysis of the state-of-the-art models. This study contributes to the ever-expanding field of drone technology by addressing critical research gaps and introducing novel approaches, but also offers practical recommendations for professionals and policymakers. The findings of this study pave the way for the integration of UAS into various applications, fostering the adoption of innovative solutions. Additionally, this research has the potential to serve as a foundational reference for future studies, encouraging collaboration and driving advancements in drone technology.</div></div>","PeriodicalId":100694,"journal":{"name":"International Journal of Cognitive Computing in Engineering","volume":"7 ","pages":"Pages 128-144"},"PeriodicalIF":0.0,"publicationDate":"2026-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145319595","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A hybrid deep learning approach based on optimized feature selection on environmental multi-disease predictive models","authors":"Selvi. S ,&nbsp;Vanathi. A","doi":"10.1016/j.ijcce.2025.11.006","DOIUrl":"10.1016/j.ijcce.2025.11.006","url":null,"abstract":"<div><div>Recent research has utilized clinical data collected over time to develop various predictive models for identifying diseases. Depending on a person's present state of health, machine learning models can find and diagnose a wide range of diseases at an early stage. However, disease forecasting still faces limitations, such as inadequate large-scale data, challenges in feature selection, and the complex structure of the models. This research introduces an innovative approach to address these issues with a multi-disease prediction model that retrieves datasets from the UCI collection, including chronic kidney disease, lung cancer, cervical cancer, Parkinson's disease, and HCC survival. The initial stage involves gathering diverse medical datasets from several reputable sources. After obtaining a set of characteristics, the optimal feature selection technique is performed using a combination of two meta-heuristic algorithms: the Salp Swarm Algorithm (SSA) and the Tuna Swarm Optimization (TSO) Algorithm. This SSATSO algorithm was specifically architected for this purpose, by combining the robust global exploration capabilities of the Salp Swarm Algorithm with the intensive local exploitation analysis of the Tuna Swarm Optimization. This dual-mechanism approach enables the algorithm to effectively traverse the complex, multi-modal search space of potential feature combinations, avoiding local optima and converging on a subset that optimally contributes to predictive accuracy. The SSATSO algorithm functions as an intelligent, adaptive filter, proactively shielding the subsequent C-SBiLSTM classifier from the detrimental effects of the curse of dimensionality. This enables the deep learning network to allocate its representational capacity exclusively to learning the complex, non-linear relationships within the most salient features, resulting in a model that is not only more accurate but also more robust and generalizable. It is important to highlight that while other standard models also exhibit some improvement with feature reduction, the proposed framework demonstrates a high-performance gradient and consistently secures the top accuracy across all subset sizes, highlighting the superior efficacy of the hybrid SSATSO approach. The proposed SSATSO<img>C-SBiLSTM framework's demonstrates clear superiority across the comprehensive suite of performance metrics, such as accuracy, precision, recall, f-score, and specificity consistently replicated across five distinct and complex disease domains, validating the framework not only in a specialized solution for a single condition, but also highly reliable platform for computational clinical decision support.</div></div>","PeriodicalId":100694,"journal":{"name":"International Journal of Cognitive Computing in Engineering","volume":"7 ","pages":"Pages 334-348"},"PeriodicalIF":0.0,"publicationDate":"2026-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145796588","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Predicting the types of physical activities using data from apple watch and fitbit devices based on machine learning (ML) methods","authors":"Qiannan Liu ,&nbsp;Yiqiao Zhang","doi":"10.1016/j.ijcce.2025.11.008","DOIUrl":"10.1016/j.ijcce.2025.11.008","url":null,"abstract":"<div><div>Apart from their communication and multimedia features, wearable health devices are highly practical for monitoring health status. The demand for wearable technologies and medical or health monitoring devices is on the rise. Because these devices are designed to collect data related to health and activity, many people want to monitor their health and vital signs. Therefore, in this study, using data from Apple Watch and Fitbit devices, two of the most important health gadgets in the field of electronic devices, a model was developed to classify and predict the type of physical activity into 6 classes. For this purpose, 8 ML frameworks, including LGBM, Extra Trees, XGB, RF, Bagging, K-Nearest Neighbors, NuSVC, and AdaBoost, were used in 2 approaches: balanced and unbalanced data. For balancing the data, the SMOTE method was used as one of the oversampling methods. The results showed that, across both Apple Watch and Fitbit devices, the LGBM classifier trained on balanced data achieved the highest evaluation indices and thus the highest accuracy.</div></div>","PeriodicalId":100694,"journal":{"name":"International Journal of Cognitive Computing in Engineering","volume":"7 ","pages":"Pages 349-364"},"PeriodicalIF":0.0,"publicationDate":"2026-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145883674","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Augmentation of comorbidity assessment of peripheral artery disease using gait biomechanics with self-organization based learning model","authors":"Nazia Sultana ,&nbsp;PK Kumar","doi":"10.1016/j.ijcce.2025.07.002","DOIUrl":"10.1016/j.ijcce.2025.07.002","url":null,"abstract":"<div><div>Peripheral artery disease (PAD) often complicates into serious complications, inclusive of Deep Vein Thrombosis (DVT), which worsens prevailing health complications, such as urinary complications and cardiac conditions related to blockage by blood clots, especially among patients with diabetes and reduced mobility. A delayed diagnosis of underdiagnosed PAD and its progression to DVT may delay treatment and worsen clinical outcome. In view of the above-mentioned challenge, the proposed work is related to the development of a self-organizing Long Short-Term Memory model to enable the early detection and comorbidity prediction of Peripheral Artery Disease patients at risk and its consequent complications, which include Deep Vein Thrombosis. In this work, we have considered the overground walking biomechanics data collected from patients with Peripheral Artery Disease, Deep Vein Thrombosis, and healthy volunteers, and extracted gait signatures at the level of hip joint angles, knee, ankle torques, powers, and ground reaction forces. The proposed self-organizing LSTM model had an accuracy of 89 % with 0.65 Matthew's Correlation Coefficient for classifying patients against these conditions after consideration of comprehensive gait characteristics. At the same time, only models with variables of GRF achieved an accuracy of 87 % with 0.64 Matthew's Correlation Coefficient. The SLSTM outperformed both Neural Networks (NN) and Random Forests (RF) across various joint-based and GRF datasets, showing superior classification accuracy (ankle, knee, and hip: &gt;0.85) and Matthews correlation coefficients (ankle: 0.70) across groups. This suggests SLSTM’s robustness in gait-based PAD classification, with highest discriminant power observed in GRF data.</div></div>","PeriodicalId":100694,"journal":{"name":"International Journal of Cognitive Computing in Engineering","volume":"7 ","pages":"Pages 365-375"},"PeriodicalIF":0.0,"publicationDate":"2026-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145924154","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Multi-Graph Neural Network attention fusion framework for emotion-aware subgraph anomaly detection in social media fake news propagation","authors":"G. Victor Daniel ,&nbsp;Chandrasekaran K. ,&nbsp;Venkatesan M. ,&nbsp;Prabhavathy P.","doi":"10.1016/j.ijcce.2025.10.011","DOIUrl":"10.1016/j.ijcce.2025.10.011","url":null,"abstract":"<div><div>Fake news on social media threatens civic trust and public safety, with emotionally charged content accelerating its spread through retweets, replies, and shares. This study addresses the challenge of detecting anomalous propagation subgraphs that reflect coordinated misinformation campaigns. We propose the Multi-Graph Neural Network Attention-based Propagation Learning for Emotion-Aware Anomaly Detection framework (hereafter referred to as MAPLE). This framework integrates sentiment features from users’ historical posts with network structures, combining multiple Graph Neural Networks (Graph Convolutional Network, Graph Attention Network, and GraphSAGE) through an attention-based fusion mechanism. Subgraph embeddings are evaluated using a One-Class Support Vector Machine for anomaly detection. Theoretical analyses establish guarantees on mutual information preservation, variance reduction, anomaly margin amplification, and entropy maximization. Experiments on the PolitiFact and GossipCop datasets show that MAPLE consistently outperforms state-of-the-art baselines, improving F1-scores by 43% and 2.17% respectively, while maintaining robustness across datasets. Unlike prior works that treat structural or sentiment cues in isolation, MAPLE provides the first unified multi-Graph Neural Network fusion framework with emotional context and theoretical underpinnings for subgraph anomaly detection.</div></div>","PeriodicalId":100694,"journal":{"name":"International Journal of Cognitive Computing in Engineering","volume":"7 ","pages":"Pages 389-399"},"PeriodicalIF":0.0,"publicationDate":"2026-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146076863","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Spatio-temporal adaptive fusion transformer for next POI recommendation","authors":"Pial Islam,&nbsp;Yeasinul Haque Sani,&nbsp;Fouzia Alam Mrittika,&nbsp;Rashedur M. Rahman","doi":"10.1016/j.ijcce.2025.10.010","DOIUrl":"10.1016/j.ijcce.2025.10.010","url":null,"abstract":"<div><div>The rapid proliferation of smartphones and location-based services has accelerated the development of intelligent systems in various fields, such as tourism, ride-sharing, delivery and shipping services, user engagement, etc. In this context, next Point of Interest (POI) recommendation plays a vital role in predicting where a user will go based on their visit sequences or check-in histories. Recent advances in sequential modeling have led to more accurate recommendations; however, capturing complex interactions across spatial, temporal, and categorical dimensions remains challenging. They are important because they affect users' desire to visit a place at a certain moment. Many current models are inadequate because they either ignore the important influence of POI categories, reduce spatial coordinates to broad region IDs, or conduct temporal analysis exclusively using recurrent neural networks. This paper presents an innovative deep-learning framework Spatio-Temporal Adaptive Fusion Transformer (STAFT), systematically addressing these challenges by seamlessly integrating continuous spatial encoding, category-aware context weighting, and learning sequential pattern. Our model employs Learnable Random Fourier Features (RFF) to transform raw spatial features into richer, high-dimensional embeddings. It also emphasizes relevant POI categories utilizing Category Aware Attention and a Temporal Convolutional Network (TCN) to efficiently capture local temporal dependencies. The Transformer Encoder architecture is used at the core to produce the final prediction. Empirical results from experimentation on real-world datasets demonstrate that our approach performs better than existing models by 4-9% on commonly used metrics.</div></div>","PeriodicalId":100694,"journal":{"name":"International Journal of Cognitive Computing in Engineering","volume":"7 ","pages":"Pages 167-188"},"PeriodicalIF":0.0,"publicationDate":"2026-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145415843","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"NeuroEvolve: A brain-inspired mutation optimization algorithm for enhancing intelligence in medical data analysis","authors":"Shailendra Pratap Singh ,&nbsp;Gyanendra Kumar ,&nbsp;Balamurugan Balusamy ,&nbsp;Nithya Rekha Sivakumar","doi":"10.1016/j.ijcce.2025.10.001","DOIUrl":"10.1016/j.ijcce.2025.10.001","url":null,"abstract":"<div><div>Medical data analysis often suffers from high dimensionality, noise, and complex non-linear patterns, causing traditional optimization-based learning methods to converge slowly and leads to suboptimal predictions. To address these challenges, we propose NeuroEvolve, a brain-inspired mutation strategy integrated into Differential Evolution (DE) that dynamically adjusts mutation factors based on feedback, enhancing both exploration and exploitation. The method was evaluated on three benchmark medical datasets – MIMIC-III, Diabetes, and Lung Cancer – using standard performance metrics including Accuracy, F1-score, Precision, Recall, and a newly defined Mean Error Correlation Coefficient (MECC). Comparative experiments with state-of-the-art evolutionary optimizers demonstrate NeuroEvolve’s superiority. For instance, on the MIMIC-III dataset, NeuroEvolve achieved an Accuracy of 94.1% and an F1-score of 91.3%, reflecting an improvement of 4.5% in Accuracy and 6.2% in F1-score over the best-performing baseline Hybrid Whale Optimization Algorithm (HyWOA). Similar improvements were consistently observed across the Diabetes and Lung Cancer datasets, confirming the robustness of the our approach. These results highlight the potential of NeuroEvolve as an effective optimization-driven learning strategy for complex medical prediction tasks.</div></div>","PeriodicalId":100694,"journal":{"name":"International Journal of Cognitive Computing in Engineering","volume":"7 ","pages":"Pages 155-166"},"PeriodicalIF":0.0,"publicationDate":"2026-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145415844","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Analysis of emotional tendencies and discourse patterns in VKontakte social comments based on Nvivo12 encoding","authors":"Jiaxing Han","doi":"10.1016/j.ijcce.2025.07.001","DOIUrl":"10.1016/j.ijcce.2025.07.001","url":null,"abstract":"<div><div>To study the emotional changes of the public during the COVID-19 epidemic, the experiment conducted an analysis of emotional tendencies and discourse patterns for comments on the VKontakte platform. The study first used Nvivo12 to classify comments on social platforms into topics, emotions, and relationship nodes. Then, a bidirectional long short-term memory network was introduced to comprehensively understand the context and classify positive and negative emotions. In addition, natural language processing toolkits were used to analyze the discourse structure of social comments, and support vector machines were used to discriminate the emotional tendencies of comments. According to the experimental analysis, during the period of rapid incidence rate increased, 27.6 % of the public exhibited positive emotional tendencies, while &lt;39.3 % exhibited negative emotional tendencies. In the following three stages, the proportion of negative emotions in the public was greater than that of positive emotions. In the fourth stage of the epidemic, comments mainly concerned the supply of medical drugs, masks, and the construction and opening of hospitals. The existing problems indicate that the epidemic has had a significant impact on public emotions, and effective measures need to be taken to alleviate the negative emotions of the public. The research results are helpful in revealing the dynamic changes of public emotions and their discourse patterns during the COVID-19 epidemic, and provide a new perspective for understanding public emotions.</div></div>","PeriodicalId":100694,"journal":{"name":"International Journal of Cognitive Computing in Engineering","volume":"7 ","pages":"Pages 1-11"},"PeriodicalIF":0.0,"publicationDate":"2026-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144686438","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An efficient and high-performance WSNs restoration algorithm for fault nodes based on FT in data aggregation scheduling","authors":"Cheng Li ,&nbsp;Guoyin Zhang","doi":"10.1016/j.ijcce.2025.05.001","DOIUrl":"10.1016/j.ijcce.2025.05.001","url":null,"abstract":"<div><div>In wireless sensor networks(WSNs), data aggregation effectively reduces network traffic, thereby reducing energy consumption and improving network life cycle. Nevertheless, in the process of data aggregation scheduling, if there are fault nodes, the data quality collected by the whole network will decline, and the network performance will decrease, even posing a threat to network security or causing network paralysis. Thus, an efficient and high-performance WSNs restoration algorithm is proposed based on fat tree(FT), which is referred to as the EPRA-FT algorithm. And our goal is to improve the universality, efficiency, and performance retention of the algorithm. Previously, we have conducted a range of the relevant researches on performance improvement of WSNs aggregation scheduling by adopting FT structure, and some successful results have been obtained. On the basis of these results, for the EPRA-FT algorithm, first and foremost, the relationship among nodes is comprehensively recorded in FT construction process. Then, fault nodes are shielded by deleting the known nodes in aggregation tree. Finally, the local reconfiguration of aggregation tree is completed quickly and efficiently. Meanwhile, the aggregation scheduling performance of the original network is maintained to the maximum extent. The feasibility and superiority of our proposed EPRA-FT algorithm are proved by simulation experiments.</div></div>","PeriodicalId":100694,"journal":{"name":"International Journal of Cognitive Computing in Engineering","volume":"6 ","pages":"Pages 508-515"},"PeriodicalIF":0.0,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143917830","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}