Egyptian Informatics Journal最新文献

{"title":"FeistelX network-based image encryption leveraging hyperchaotic fusion and extended DNA coding","authors":"Kristiawan Nugroho ,&nbsp;De Rosal Ignatius Moses Setiadi ,&nbsp;Eri Zuliarso ,&nbsp;Aceng Sambas ,&nbsp;Omar Farooq","doi":"10.1016/j.eij.2025.100716","DOIUrl":"10.1016/j.eij.2025.100716","url":null,"abstract":"<div><div>The rising frequency of cyberattacks has heightened the need for more secure and efficient image encryption techniques. Traditional chaotic and DNA-based methods often struggle with limited key space, low diffusion efficiency, or vulnerability to statistical attacks, especially when handling large or high-dimensional image data. This study introduces an image encryption technique that integrates the FeistelX Network with extended DNA cryptography and two distinct two-dimensional hyperchaotic maps, namely the two-dimensional symbolic chaotic map (2D-SCM) and the two-dimensional hyperchaotic exponential adjusted Logistic and Sine map (2D-HELS), to bolster data security. The proposed method synergizes three key components: the FeistelX Network offers a robust encryption framework with bijectivity ensured by property H; the extended DNA cryptography expands the key space and minimizes pixel correlation through advanced DNA operations; and the two hyperchaotic maps generate highly intricate chaotic sequences, ensuring greater randomness and resilience. Compared to existing schemes, the proposed method demonstrates improved diffusion, randomness, and resistance to statistical attacks. Experimental results show that this method achieves high-security indicators, with Chi-square values consistently below the critical threshold, average entropy values of 7.9994, and UACI and NPCR metrics remaining within the optimal theoretical ranges. Moreover, the method passed all sixteen NIST randomness tests with an average p-value of 0.6278. It demonstrated resilience to noise and data loss with PSNR values above 18 dB under attack scenarios. This combination of FeistelX structure, extended DNA operations, and dual hyperchaotic maps offers a novel and effective solution for enhancing image encryption security beyond traditional approaches.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"31 ","pages":"Article 100716"},"PeriodicalIF":5.0,"publicationDate":"2025-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144322339","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Named entity recognition using Bi-LSTM model with pointer cascade conditional random field for selecting high-profit products","authors":"C. Gayathri ,&nbsp;Dr. R. Samson Ravindran","doi":"10.1016/j.eij.2025.100703","DOIUrl":"10.1016/j.eij.2025.100703","url":null,"abstract":"<div><div>Named entity recognition (NER) refers to recognizing objects mentioned in texts and is considered one of the most fundamental tasks in natural language processing. The authentication of named entities is not merely a matter of extracting information independently. The rise of this sector has benefited from rapid growth, especially in the e-commerce sector; numerous reviews are published that reflect consumer sentiments on different aspects of products and services such as quality, price, and more.A critical challenge lies in improving the accuracy and robustness of NER systems to address issues such as ambiguous contexts, intricate sentence structures, and domain-specific variations. Previous works on NER usually use conventional machine learning methods. However, there is still a need to improve the accuracy of identifying entities. To accomplish this goal, this work proposes a pointer cascade conditional random field-based named entity recognition procedure. A word embedding approach is initially applied to segment the word for further processing. Word vectors are provided as input to a bidirectional LSTM (Bi-LSTM) model, which extracts features from sentence or word vectors. To improve the performance of BiLSTM, a pointer network is used to generate pointer sequences for the elements of the input array. After features are extracted, the Cascade Conditional Random Field (CCRF) layer checks tag validity by learning the correlation between tags. A Python 3.7 framework is used to implement the proposed model. According to the results of the experiments, this work achieves a high accuracy of 98.54 %.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"31 ","pages":"Article 100703"},"PeriodicalIF":5.0,"publicationDate":"2025-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144312957","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"MalwareNet: an intelligent malware detection and classification using advanced extreme leaning machine in edge computing environment","authors":"P. Shailaja ,&nbsp;Thanveer Jahan ,&nbsp;Karramreddy Sharmila ,&nbsp;P. Bharath Siva Varma ,&nbsp;Swetha Arra ,&nbsp;Pala Mahesh Kumar","doi":"10.1016/j.eij.2025.100714","DOIUrl":"10.1016/j.eij.2025.100714","url":null,"abstract":"<div><div>Malware continues to wreak havoc on global digital ecosystems, with companies facing an average financial loss of $4.35 million per data breach in recent years. At the same time, individual users suffer from identity theft, affecting over 1.1 billion personal records annually. Existing malware detection systems often struggle with high latency in centralized cloud environments and fail to generalize across diverse malware variants generated by edge devices. To address these challenges, this work introduces MalwareNet, a novel multiclass malware detection network designed specifically for edge computing environments. MalwareNet innovatively processes data directly on edge devices, enabling real-time detection and classification with minimal latency and enhanced data privacy. The system employs a robust preprocessing pipeline to clean raw data, followed by Independent Component Analysis (ICA) to extract discriminative features while reducing dataset dimensionality. A Hybrid Wrapper-Filter (HWF) feature selection method optimizes feature subsets by integrating wrapper and filter techniques, ensuring compatibility with the chosen machine-learning classifier to maximize classification accuracy. The Extreme Learning Machine (ELM), selected for its rapid training and strong generalization, classifies malware into distinct categories, effectively identifying threats in edge settings. By combining edge-based processing, advanced feature engineering, and efficient classification, MalwareNet offers a scalable and reliable solution, significantly advancing malware detection capabilities for resource-constrained environments and providing a foundation for future adaptive security systems. Experimental evaluations on a large-scale malware dataset demonstrate the effectiveness of the proposed approach with an accuracy of 99.7 %, and F-measure of 99.55 %. The system also achieves high Jaccard index with an increment of 2.63 % in detecting and classifying malware, providing reliable security measures in edge computing environments.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"31 ","pages":"Article 100714"},"PeriodicalIF":5.0,"publicationDate":"2025-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144322338","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A federated supply chain finance risk control method based on personalized differential privacy","authors":"Chao Ma,&nbsp;Haiyu Zhao,&nbsp;Kaiqi Zhang,&nbsp;Luogang Zhang,&nbsp;Hai Huang","doi":"10.1016/j.eij.2025.100704","DOIUrl":"10.1016/j.eij.2025.100704","url":null,"abstract":"<div><div>With the rapid development of supply chain finance, effectively managing risks while safeguarding participant data privacy has become a critical area of research. However, existing traditional risk control models predominantly rely on centralized data processing, which leads to the phenomenon of ”data silos,” hindering the flow and sharing of information. Furthermore, the significant privacy risks associated with centralized processing restrict collaboration among financial institutions, exacerbating the challenges of risk management. In this context, this study proposes a federated risk control method for supply chain finance based on personalized differential privacy optimization. This approach introduces a personalized differential privacy mechanism, enabling different institutions to collaboratively optimize model parameters without directly exchanging sensitive data. This methodology not only effectively safeguards data privacy but also enhances the overall performance of risk control, facilitating multi-party collaboration. Experimental results indicate that, compared to traditional centralized risk control models and other privacy protection methods, the proposed solution demonstrates favorable outcomes in terms of predictive accuracy and model performance while adhering to data privacy protection requirements. This research lays a theoretical foundation for the future development of safer and more efficient cross-institutional risk control systems and provides new insights and technical support for innovative risk management in the field of supply chain finance.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"31 ","pages":"Article 100704"},"PeriodicalIF":5.0,"publicationDate":"2025-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144312956","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Analysis of social networks content to identify fake news using stacked combination of deep neural networks","authors":"Yujie Li ,&nbsp;Yushui Xiao ,&nbsp;Yong Huang ,&nbsp;Rui Ma","doi":"10.1016/j.eij.2025.100707","DOIUrl":"10.1016/j.eij.2025.100707","url":null,"abstract":"<div><div>In today’s fast-paced world, the unprecedented expansion of social networks and the huge volume of information has made automatic detection of fake news an undeniable necessity. The dissemination of fake news and misinformation can have a devastating impact on public opinion and social decision-making. This challenge requires new and powerful approaches in the fields of deep learning and natural language processing to accurately and quickly identify fake news and prevent its dissemination. For that purpose, this current work presents a new and efficient solution to detecting and spotting spurious news on social media. This method, through deep text content analysis and the employment of advanced deep learning techniques, aims to provide an expansive and accurate response to solve this problem. The proposed method consists of three determining steps: 1) The input data is initially prepared for the next steps using preprocessing techniques. This is done through noise removal, text normalization, and data conversion into a format that can be processed by deep learning models. 2) A hybrid method is then used to extract text features, which is a combination of a list of statistical features (e.g., text length, word count, and links), GloVe-based semantic features (to represent the word relationships), and Character N-Grams (CNG) (to improve misspelling and linguistic anomaly robustness). 3) Finally, for each set of features, a particular deep model is trained to predict based on each component. Specifically, a Multilayer Perceptron (MLP) model is used for statistical feature analysis, and Convolutional Neural Network (CNN) models are used for GloVe and CNG features. Both models generate individual predictions from the input features presented to them, and the predicted labels and the posterior probability vector for each of the models are combined to output a vector to be forwarded to the <em>meta</em>-learner (a MLP model). By learning patterns in the combinations of outputs and the probability vectors of the individual base models, the MLP model can correctly identify fake news or real news. Experimental results conducted on two authentic datasets, GossipCop and Politifact, show that our proposed method achieves 99.45 % and 97.40 % accuracies, respectively. This achievement indicates the very good and effective performance of our method in detecting fake news on both datasets.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"30 ","pages":"Article 100707"},"PeriodicalIF":5.0,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144196380","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Cancelable finger vein authentication using multidimensional scaling based on deep learning","authors":"Mohamed Hammad ,&nbsp;Mohammed ElAffendi ,&nbsp;Ahmed A. Abd El-Latif","doi":"10.1016/j.eij.2025.100708","DOIUrl":"10.1016/j.eij.2025.100708","url":null,"abstract":"<div><div>In the field of identity verification and identification, biometrics has evolved as a reliable approach for identifying individuals based on their unique physical or behavioral characteristics. The utilization of finger vein authentication has generated significant attention as a biometric modality owing to its strong resilience, resistance against spoofing attacks, and consistent patterns. In this work, we proposed a novel cancelable finger vein authentication system using multidimensional scaling (MDS) based on deep learning. Our method addressed the limitations of previous biometric authentication systems by integrating MDS with a lightweight convolutional neural network (CNN) model for feature extraction. The cancelable approach ensured privacy and security by generating distinct templates for each user. We evaluated our system on <em>three</em> publicly available datasets for finger veins using various performance metrics, including accuracy, precision, recall, and equal error rate (EER). The results demonstrated the effectiveness of our method, which achieved high accuracy, low error rates, and strong performance in diversity and irreversibility tests. Additionally, our system maintained high authentication accuracy while preserving user privacy, making it suitable for practical applications in biometric authentication.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"30 ","pages":"Article 100708"},"PeriodicalIF":5.0,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144220953","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Innovative quantum techniques for improving system performance in cloud computing","authors":"Naher M. Alsafri ,&nbsp;Ahmed Y. Hamed ,&nbsp;A. Mindil ,&nbsp;M.R. Hassan","doi":"10.1016/j.eij.2025.100710","DOIUrl":"10.1016/j.eij.2025.100710","url":null,"abstract":"<div><div>Effective task scheduling is pivotal for optimizing the performance of cloud computing services, particularly to minimize execution time and enhance resource utilization. Traditional approaches often focus on single-objective metrics, such as task completion time, or fail to address the intricate interdependencies between multiple objectives. To overcome these limitations, we introduce QISPF, a novel multi-objective task scheduling algorithm that combines genetic algorithms with innovative quantum techniques. QISPF is designed to achieve an optimal task distribution by addressing key performance metrics makespan, scheduling length, throughput, resource utilization, energy consumption, and load balancing, through a unified measure known as system performance. QISPF leverages quantum techniques to enhance the traditional genetic algorithm framework by incorporating principles from quantum mechanics, such as probabilistic quantum encoding and superposition. The simulations were conducted for two cases. The first had 100 tasks and anything from 10 to 50 virtual machines. Furthermore, in the second case, there were a certain number of virtual machines (VMs), with the number of tasks ranging from 500 to 1000. The simulation results demonstrated the scheduling efficiency of QISPF compared to the G-MOTSA, ETVMC, TSACS, and ACO algorithms. QISPF offers a more powerful approach to exploring and exploiting the solution space. This novel method allows for a richer representation of potential solutions and improves the algorithm’s ability to find high-quality solutions in complex problem landscapes.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"30 ","pages":"Article 100710"},"PeriodicalIF":5.0,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144229763","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Machine learning models for enhanced stroke detection and prediction","authors":"Shilpa Bajaj ,&nbsp;Manju Bala ,&nbsp;Mohit Angurala","doi":"10.1016/j.eij.2025.100705","DOIUrl":"10.1016/j.eij.2025.100705","url":null,"abstract":"<div><div>Stroke detection plays a vital role in medical diagnostics, where timely and accurate identification can improve patient outcomes. This research evaluates the performance of three machine learning models—OzNet-mRMR-NB, Logistics Regression, and an Ensemble CNN—using medical images for stroke prediction. The OzNet-mRMR-NB model integrates VGG19 for feature extraction, mRMR for feature selection, and Naive Bayes for classification, while Logistic Regression processes flattened feature vectors. The Ensemble CNN, leveraging ResNet and VGG19, outperforms the other models with a testing accuracy of 92.43 %, an AUC score of 0.92, precision of 0.93, and an F1-score of 0.92. Additionally, recall for both the Ensemble and OzNet models was 0.93, highlighting the Ensemble model’s capacity to sustain a robust balance between specificity and sensitivity. These results highlight the advantages of combining diverse CNN architectures for improved accuracy and generalizability. This research advances automated stroke detection, with potential clinical applications for timely and informed decision-making. Future work will refine the ensemble approach for broader clinical adoption across diverse patient populations.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"30 ","pages":"Article 100705"},"PeriodicalIF":5.0,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144241315","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Emotion recognition in panoramic audio and video virtual reality based on deep learning and feature fusion","authors":"Siqi Guo,&nbsp;Mian Wu,&nbsp;Chunhui Zhang,&nbsp;Ling Zhong","doi":"10.1016/j.eij.2025.100697","DOIUrl":"10.1016/j.eij.2025.100697","url":null,"abstract":"<div><div>Virtual reality technology has been widely applied in various fields of society, and its content emotion recognition has received much attention. The recognition of emotions in virtual reality content can be employed to regulate emotional states in accordance with the emotional content, to treat mental illness and to assess psychological cognition. Nevertheless, the current research on emotion induction and recognition of virtual reality scenes lacks scientific and quantitative methods for establishing the mapping relationship between virtual reality scenes and emotion labels. Furthermore, the associated methods lack clarity regarding image feature extraction, which contributes to the diminished accuracy of emotion recognition in virtual reality content. To solve the current issue of inaccurate emotion recognition in virtual reality content, this study combines convolutional neural networks and long short-term memory. The attention mechanism and multi-modal feature fusion are introduced to improve the speed of feature extraction and convergence. Finally, an improved algorithm-based emotion recognition model for panoramic audio and video virtual reality is proposed. The average accuracy of the proposed algorithm, XLNet-BIGRU-Attention algorithm, and CNN-BiLSTM algorithm was 98.87%, 90.25%, and 86.21%, respectively. The average precision was 98.97%, 97.24% and 97.69%, respectively. The proposed algorithm was significantly superior to the comparison algorithm. A performance comparison was conducted between panoramic audio and video virtual reality emotion recognition models based on the improved algorithm. The improved algorithm’s the mean square error is 0.17 and mean absolute error is 0.19, obviously better than other comparison models. In the analysis of visual classification results, the proposed model has the best classification aggregation effect and is significantly superior to other models. Therefore, the improved algorithm and the panoramic audio and video virtual reality emotion recognition model based on the improved algorithm have good effectiveness and practical value.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"30 ","pages":"Article 100697"},"PeriodicalIF":5.0,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144241316","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Intrusion detection system framework for cyber-physical systems","authors":"Shafiq ur Rehman ,&nbsp;Hisham Alhulayyil ,&nbsp;Taher Alzahrani ,&nbsp;Hatoon AlSagri ,&nbsp;Muhammad U. Khalid ,&nbsp;Volker Gruhn","doi":"10.1016/j.eij.2024.100600","DOIUrl":"10.1016/j.eij.2024.100600","url":null,"abstract":"<div><div>Cyber-Physical Systems (CPS) have become integral components across diverse sectors, including autonomous vehicle systems, healthcare, power distribution, and manufacturing. These systems leverage physical components enhanced with intelligent capabilities, enabling autonomous functionality and increased efficiency. Security is a critical concern for CPS due to their close integration with essential infrastructure, where failures can have severe consequences for both the physical environment and human lives. Intrusion Detection Systems (IDS) can be a vital tool for secure CPS, detecting and alerting against threats such as malicious activities. However, conventional IDS designs are often inadequate for CPS environments, typically focusing solely on the network (Network-based Intrusion Detection System or NIDS) or application layer (Host-based Intrusion Detection System or HIDS), while neglecting the physical layer. Therefore, this research proposes a novel IDS framework that employs a hybrid detection approach, along with comprehensive guidelines for intrusion detection specifically tailored to CPS. This initiative contributes towards establishing a cohesive IDS framework for CPS, empowering practitioners in navigating this domain and crafting bespoke intrusion detection solutions. The proposed approach has been rigorously evaluated through a comparative analysis of different methodologies, demonstrating the effectiveness of the guidelines and requirements in addressing all relevant security aspects for IDS design. This research provides CPS practitioners and researchers with actionable guidelines designed to effectively enhance the security posture of their systems. By implementing these guidelines, they can better protect against threats and mitigate their potential consequences, thereby contributing to the security of Industry 4.0. This proactive approach not only secure critical infrastructure but also fosters a more resilient and secure operational environment in the face of evolving cyber threats.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"30 ","pages":"Article 100600"},"PeriodicalIF":5.0,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144212211","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}