Egyptian Informatics Journal最新文献

{"title":"Evolutionary approach for composing a thoroughly optimized ensemble of regression neural networks","authors":"Lazar Krstic,&nbsp;Milos Ivanovic,&nbsp;Visnja Simic,&nbsp;Boban Stojanovic","doi":"10.1016/j.eij.2024.100581","DOIUrl":"10.1016/j.eij.2024.100581","url":null,"abstract":"<div><div>The paper presents the GeNNsem (<strong>Ge</strong>netic algorithm A<strong>NN</strong>s en<strong>sem</strong>ble) software framework for the simultaneous optimization of individual neural networks and building their optimal ensemble. The proposed framework employs a genetic algorithm to search for suitable architectures and hyperparameters of the individual neural networks to maximize the weighted sum of accuracy and diversity in their predictions. The optimal ensemble consists of networks with low errors but diverse predictions, resulting in a more generalized model. The scalability of the proposed framework is ensured by utilizing micro-services and Kubernetes batching orchestration. GeNNsem has been evaluated on two regression benchmark problems and compared with related machine learning techniques. The proposed approach exhibited supremacy over other ensemble approaches and individual neural networks in all common regression modeling metrics. Real-world use-case experiments in the domain of hydro-informatics have further demonstrated the main advantages of GeNNsem: requires the least training sessions for individual models when optimizing an ensemble; networks in an ensemble are generally simple due to the regularization provided by a trivial initial population and custom genetic operators; execution times are reduced by two orders of magnitude as a result of parallelization.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"28 ","pages":"Article 100581"},"PeriodicalIF":5.0,"publicationDate":"2024-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143181857","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Improving healthy food recommender systems through heterogeneous hypergraph learning","authors":"Jing Wang ,&nbsp;Jincheng Zhou ,&nbsp;Muammer Aksoy ,&nbsp;Nidhi Sharma ,&nbsp;Md Arafatur Rahman ,&nbsp;Jasni Mohamad Zain ,&nbsp;Mohammed J.F. Alenazi ,&nbsp;Aliyeh Aminzadeh","doi":"10.1016/j.eij.2024.100570","DOIUrl":"10.1016/j.eij.2024.100570","url":null,"abstract":"<div><div>Recommender systems in health-conscious recipe suggestions have evolved rapidly, particularly with the integration of both homogeneous and heterogeneous graphs. However, incorporating IoT devices into healthcare, such as wearable fitness trackers and smart nutrition scales, presents new challenges. These devices generate vast amounts of dynamic, personalized data, which traditional Graph Neural Network (GNN) models — limited to simple pairwise connections — fail to capture effectively. For example, IoT sensors tracking daily nutrient intake require complex, multi-faceted analysis that traditional methods struggle to handle. To overcome these limitations, researchers have employed hypergraphs, which capture higher-order relationships among nodes, such as user–food and ingredient interactions. Traditional methods using static weights in the Laplacian hypergraph, inspired by homogeneous graph techniques, often fail to account for users’ evolving health interests. Our study introduces a novel approach for recommending healthy foods by leveraging user–food and food-ingredient hyperedges, integrating both convolution and attention-based hypergraph mechanisms to dynamically adjust weights based on user similarities. Unlike previous methods, we convert the heterogeneous hypergraph into a homogeneous space, using a unified loss function to generate recommendations that adapt to individual users’ changing dietary preferences. The model is evaluated on five metrics — AUC, NDCG, Precision, Recall, and F1-score — and shows superior performance compared to existing models on two real-world food datasets, Allrecipes and Food.com. Our results demonstrate significant improvements in recommendation accuracy and personalization, showcasing the system’s effectiveness in integrating IoT data for more responsive, health-focused food suggestions.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"28 ","pages":"Article 100570"},"PeriodicalIF":5.0,"publicationDate":"2024-11-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142722590","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Efficient GDD feature approximation based brain tumour classification and survival analysis model using deep learning","authors":"M. Vimala ,&nbsp;SatheeshKumar Palanisamy ,&nbsp;Sghaier Guizani ,&nbsp;Habib Hamam","doi":"10.1016/j.eij.2024.100577","DOIUrl":"10.1016/j.eij.2024.100577","url":null,"abstract":"<div><div>The problem of brain tumor classification (BTC) has been approached with several methods and uses different features obtained from MRI brain scans. However, they suffer from achieving higher performance in BTC and produce poor performance with a higher false ratio. A convolutional neural network (CNN) based on BTC and a survival analysis model based on GDD (growth distribution depth) are presented. Initially, an adaptive median filter (AMF) is used to preprocess the MRI images in order to lower the amount of noise in the images. Secondly, in order to calculate the GDD value, the texture, shape, and gradient characteristics are extracted. Third, CNN is used to train the retrieved features based on the labels that were found. In the classification, the GDD features extracted are used to measure TSF (Tumor Support Factor) in each of them. The neurons of the network measure the value of tumor weight (TW) to perform classification. Additionally, the technique evaluates a patient’s survival and calculates the survival rate based on the TSF value of the growth characteristic. The multi-layer perceptron allows the computation of TW and supports the efficient performance of classification. The proposed method improves tumor classification performance by up to 97%.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"28 ","pages":"Article 100577"},"PeriodicalIF":5.0,"publicationDate":"2024-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142722589","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The best angle correction of basketball shooting based on the fusion of time series features and dual CNN","authors":"Meicai Xiao","doi":"10.1016/j.eij.2024.100579","DOIUrl":"10.1016/j.eij.2024.100579","url":null,"abstract":"<div><div>The best shooting angle correction of basketball based on intelligent image analysis is an important branch of the development of intelligent sports. However, the current method is limited by the variability of the shape base, ignoring dynamic features and visual information, and there are some problems in the process of feature extraction and correction of related actions. This paper proposes a method to correct the best shooting angle of basketball based on the fusion of time series characteristics and dual CNN. Segmenting the shooting video, taking the video frame as the input of the key node extraction network of the shooting action, obtaining the video frame with the sequence information of the bone points, extracting the continuous T-frame video stack from it, and inputting it into the spatial context feature extraction network in the shooting posture prediction model based on dual stream CNN (MobileNet V3 network with multi-channel attention mechanism fusion module), extract the space context features of shooting posture; The superimposed optical flow graph of continuous video frames containing sequence information of bone points is input into the time convolution network (combined with Bi-LSTM network of multi-channel attention mechanism fusion module), extract the skeleton temporal sequence features during the shooting movement, using the spatial context features and skeleton temporal sequence features extracted from the feature fusion module, and realizing the prediction of shooting posture through Softmax according to the fusion results, calculate the shooting release speed under this attitude, solve the shooting release angle, and complete the correction of the best shooting release angle by comparing with the set conditions. The experimental results show that this method can achieve the best shooting angle correction, and the training learning rate is 0.2 × 10–3, training loss is about 0.05; MPJPE and MPJVE indicators are the lowest, and Top-1 indicators are the highest; The shooting percentage is about 95 %.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"28 ","pages":"Article 100579"},"PeriodicalIF":5.0,"publicationDate":"2024-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142706572","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multi-factor evaluation of clustering methods for e-commerce application","authors":"Adam Wasilewski ,&nbsp;Krzysztof Juszczyszyn ,&nbsp;Vera Suryani","doi":"10.1016/j.eij.2024.100562","DOIUrl":"10.1016/j.eij.2024.100562","url":null,"abstract":"<div><div>This research aimed to investigate the application of Vlse Kriterijumska Optimizacija I Kompromisno Resenje (VIKOR) and Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) multi-criteria decision-making to select the optimal clustering for e-commerce customer segmentation. In this context, clustering as an unsupervised machine learning method offered a way to overcome the limitations of traditional grouping, particularly by providing the ability to capture the diverse needs of consumers. A total of five different clustering methods were considered based on the behavioral data of e-commerce customers. Even though the analyzed algorithms were well-known and widely used, the comprehensive and multidirectional comparison was not trivial. Selected approaches were evaluated on the basis of twelve indicators (decision criteria), divided into four characteristics that take into account both the out-of-context aspects of clustering and the requirements arising from the context of using the clustering results. The results showed consistent outcomes from both analyzed Multi-Criteria Decision Methods, with some notable differences. The methods obtained the same ranking of the top three clustering algorithms (K-median - BIRCH - K-means). However, the TOPSIS and VIKOR sensitivity analysis recommended K-means in 87% of the cases and 60% of the variants verified, respectively. The parameterization of the decision factors had a significant impact on the final ranking of clustering options. This research demonstrated the practical application of the decision methods in selecting the best clustering for multivariate user interfaces to improve personalization in e-commerce.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"28 ","pages":"Article 100562"},"PeriodicalIF":5.0,"publicationDate":"2024-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142706571","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A study of Isual perceptual target monitoring in graphic design based on Multi-Task structured learning and interaction mapping","authors":"Jingchao Liu ,&nbsp;Yang Zhang ,&nbsp;Jing Wang","doi":"10.1016/j.eij.2024.100576","DOIUrl":"10.1016/j.eij.2024.100576","url":null,"abstract":"<div><div>In graphic design, many materials come from images and videos, but the current visual target analysis still suffers from the disadvantages of poor results and not being able to understand the semantic information required by graphic design. In order to solve the above problems, this study builds a visual perception target monitoring network model by combining multi-task structured learning and interaction mapping detection methods, and based on the combined detection method. The study first analyses the target detection effect of the combined detection method, and the results show that compared with other methods, the ROC curve area of the method used in this paper is larger and the accuracy is higher, up to 96.45 %, and the maximum accuracy of the detection method is 90.00 %. Then the target tracking effect of the combined detection method is analysed, and the average success rate of the proposed method in multi-target tracking is maximum 99.69 %. Finally, the model’s effectiveness in target classification and identification is analysed, and the results show that the classification error rate of the network model based on the detection method is 4.99 %, which is lower than other models. From the above results, it can be seen that the visual perception target monitoring network model based on multi-task structured learning and interaction mapping detection method proposed in the study can achieve visual target perception and has certain application value in graphic design.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"28 ","pages":"Article 100576"},"PeriodicalIF":5.0,"publicationDate":"2024-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142706570","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An improved multiobjective evolutionary algorithm for time-dependent vehicle routing problem with time windows","authors":"Jia-ke Li ,&nbsp;Jun-qing Li ,&nbsp;Ying Xu","doi":"10.1016/j.eij.2024.100574","DOIUrl":"10.1016/j.eij.2024.100574","url":null,"abstract":"<div><div>Time-dependent vehicle routing problem with time windows (TDVRPTW) is a pivotal problem in logistics domain. In this study, a special case of TDVRPTW with temporal-spatial distance (TDVRPTW-TSD) is investigated, which objectives are to minimize the total travel time and maximize customer satisfaction while satisfying the vehicle capacity. To address it, an improved multiobjective evolutionary algorithm (IMOEA) is developed. In the proposed algorithm, a hybrid initialization strategy with two efficient heuristics considering temporal-spatial distance is designed to generate high-quality and diverse initial solutions. Then, two crossover operators are devised to broaden the exploration space. Moreover, an efficient local search heuristic combing the adaptive large neighborhood search (ALNS) and the variable neighborhood descent (VND) is developed to improve the exploration capability. Finally, detailed comparisons with several state-of-the-art algorithms are tested on a set of instances, which verify the efficiency and effectiveness of the proposed IMOEA.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"28 ","pages":"Article 100574"},"PeriodicalIF":5.0,"publicationDate":"2024-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142658097","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Distributed hybrid flowshop scheduling with consistent sublots under delivery time windows: A penalty lot-assisted iterated greedy algorithm","authors":"Jinli Liu ,&nbsp;Yuyan Han ,&nbsp;Yuting Wang ,&nbsp;Yiping Liu ,&nbsp;Biao Zhang","doi":"10.1016/j.eij.2024.100566","DOIUrl":"10.1016/j.eij.2024.100566","url":null,"abstract":"&lt;div&gt;&lt;div&gt;Integrating the delivery time windows into the distributed hybrid flow shop scheduling contributes to ensuring the timely delivery of products and enhancing customer satisfaction. In view of this, this study focuses on distributed hybrid flowshop scheduling with consistent sublots under the delivery time windows constraint, denoted as &lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;mi&gt;DH&lt;/mi&gt;&lt;msub&gt;&lt;mi&gt;F&lt;/mi&gt;&lt;mi&gt;m&lt;/mi&gt;&lt;/msub&gt;&lt;mrow&gt;&lt;mo&gt;|&lt;/mo&gt;&lt;msub&gt;&lt;mrow&gt;&lt;mi&gt;lot&lt;/mi&gt;&lt;/mrow&gt;&lt;mrow&gt;&lt;mi&gt;cs&lt;/mi&gt;&lt;/mrow&gt;&lt;/msub&gt;&lt;mo&gt;|&lt;/mo&gt;&lt;/mrow&gt;&lt;mi&gt;ε&lt;/mi&gt;&lt;mrow&gt;&lt;mfenced&gt;&lt;mrow&gt;&lt;mi&gt;T&lt;/mi&gt;&lt;mi&gt;W&lt;/mi&gt;&lt;mi&gt;E&lt;/mi&gt;&lt;mi&gt;T&lt;/mi&gt;&lt;mo&gt;/&lt;/mo&gt;&lt;mi&gt;D&lt;/mi&gt;&lt;mi&gt;T&lt;/mi&gt;&lt;mi&gt;W&lt;/mi&gt;&lt;/mrow&gt;&lt;/mfenced&gt;&lt;/mrow&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt;. However, there exist some challenges of problem model modeling and algorithmic design for the problem to be addressed. Therefore, we first construct a mixed integer linear programming (MILP) model tailored to &lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;mi&gt;DH&lt;/mi&gt;&lt;msub&gt;&lt;mi&gt;F&lt;/mi&gt;&lt;mi&gt;m&lt;/mi&gt;&lt;/msub&gt;&lt;mrow&gt;&lt;mo&gt;|&lt;/mo&gt;&lt;msub&gt;&lt;mrow&gt;&lt;mi&gt;lot&lt;/mi&gt;&lt;/mrow&gt;&lt;mrow&gt;&lt;mi&gt;cs&lt;/mi&gt;&lt;/mrow&gt;&lt;/msub&gt;&lt;mo&gt;|&lt;/mo&gt;&lt;/mrow&gt;&lt;mi&gt;ε&lt;/mi&gt;&lt;mrow&gt;&lt;mfenced&gt;&lt;mrow&gt;&lt;mi&gt;T&lt;/mi&gt;&lt;mi&gt;W&lt;/mi&gt;&lt;mi&gt;E&lt;/mi&gt;&lt;mi&gt;T&lt;/mi&gt;&lt;mo&gt;/&lt;/mo&gt;&lt;mi&gt;D&lt;/mi&gt;&lt;mi&gt;T&lt;/mi&gt;&lt;mi&gt;W&lt;/mi&gt;&lt;/mrow&gt;&lt;/mfenced&gt;&lt;/mrow&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt; with the aim of minimizing the total weighted earliness and tardiness (&lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;mi&gt;TWET&lt;/mi&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt;). Additionally, we introduce a penalty lot-assisted iterated greedy (&lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;mi&gt;P&lt;/mi&gt;&lt;mi&gt;L&lt;/mi&gt;&lt;mi&gt;_&lt;/mi&gt;&lt;mi&gt;I&lt;/mi&gt;&lt;mi&gt;G&lt;/mi&gt;&lt;mi&gt;_&lt;/mi&gt;&lt;mi&gt;I&lt;/mi&gt;&lt;mi&gt;T&lt;/mi&gt;&lt;mi&gt;I&lt;/mi&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt;) and idle time insertion to coincide better with delivery time windows, in which a delivery-time-based multi-rule NEH, an adaptive insertion-based reconstruction based on the changing of the delivery status, a trilaminar penalty lot-assisted local search, and an elitist list-based acceptance criterion are designed to save convergence time and reduce the late deliveries attempts. Lastly, we also introduce a completely new method to generate delivery time windows and create 400 distinct instances. Based on the average results from five runs of 400 instances, &lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;mi&gt;P&lt;/mi&gt;&lt;mi&gt;L&lt;/mi&gt;&lt;mi&gt;_&lt;/mi&gt;&lt;mi&gt;I&lt;/mi&gt;&lt;mi&gt;G&lt;/mi&gt;&lt;mi&gt;_&lt;/mi&gt;&lt;mi&gt;I&lt;/mi&gt;&lt;mi&gt;T&lt;/mi&gt;&lt;mi&gt;I&lt;/mi&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt; demonstrates improvements of 59.0 %, 72.3 %, 76.9 %, and 25.5 % compared to &lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;mi&gt;HIGT&lt;/mi&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt;, &lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;mi&gt;DABC&lt;/mi&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt;, &lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;mi&gt;CVND&lt;/mi&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt;, and &lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;mi&gt;I&lt;/mi&gt;&lt;mi&gt;G&lt;/mi&gt;&lt;mi&gt;_&lt;/mi&gt;&lt;mi&gt;M&lt;/mi&gt;&lt;mi&gt;R&lt;/mi&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt;, respectively. When considering the minimum values from each instance, &lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;mi&gt;P&lt;/mi&gt;&lt;mi&gt;L&lt;/mi&gt;&lt;mi&gt;_&lt;/mi&gt;&lt;mi&gt;I&lt;/mi&gt;&lt;mi&gt;G&lt;/mi&gt;&lt;mi&gt;_&lt;/mi&gt;&lt;mi&gt;I&lt;/mi&gt;&lt;mi&gt;T&lt;/mi&gt;&lt;mi&gt;I&lt;/mi&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt; exhibits enhancements of 59.4 %, 71.8 %, 74.9 %, and 25.4 % over &lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;mi&gt;HIGT&lt;/mi&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt;, &lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;mi&gt;DABC&lt;/mi&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt;, &lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;mi&gt;CVND&lt;/mi&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt;, and &lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;mi&gt;I&lt;/mi&gt;&lt;mi&gt;G&lt;/","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"28 ","pages":"Article 100566"},"PeriodicalIF":5.0,"publicationDate":"2024-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142658095","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Cyber epidemic spread forecasting based on the entropy-extremal dynamic interpretation of the SIR model","authors":"Viacheslav Kovtun ,&nbsp;Krzysztof Grochla ,&nbsp;Mohammed Al-Maitah ,&nbsp;Saad Aldosary ,&nbsp;Tetiana Gryshchuk","doi":"10.1016/j.eij.2024.100572","DOIUrl":"10.1016/j.eij.2024.100572","url":null,"abstract":"<div><div>The spread of a cyber epidemic at an early stage is an uncertain process characterized by a small amount of statistically unreliable data. Nonlinear dynamic models, most commonly the SIR model, are widely used to describe such processes. The description of the studied process obtained using this model is sensitive to the initial conditions set and the quality of tuning the controlled parameters based on the results of operational observations, which are inherently uncertain. This article proposes a transition to a stochastic interpretation of the controlled parameters of the SIR model and the introduction of additional stochastic parameters that represent the variability of operational data measurements. The process of estimating the probability density functions of these parameters and noises is implemented as a strict optimization problem. The resulting mathematical apparatus is generalized in the form of two versions of the entropy-extremal adaptation of the SIR model, which are applied to forecast the spread of a cyber epidemic. The first version is focused on estimating the SIR model parameters based on operational data. In contrast, the second version focuses on stochastic modelling of the transmission rate indicator and its impact on forecasting the studied process. The forecasting result represents the average trajectory from the set of trajectories obtained using the authors’ models, which characterize the dynamics of compartment <em>I</em>. The experimental part of the article compares the classical Least Squares method with the authors’ entropy-extremal approach for estimating the SIR model parameters based on etalon data on the spread of the most threatening categories of malware cyber epidemics. The empirical results are characterized by a significant reduction in the Mean Absolute Percentage Error regarding the etalon data over the prediction interval, which proves the adequacy of the proposed approach.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"28 ","pages":"Article 100572"},"PeriodicalIF":5.0,"publicationDate":"2024-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142658096","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"BACAD: AI-based framework for detecting vertical broken access control attacks","authors":"Ahmed Anas ,&nbsp;Ayman A. Alhelbawy ,&nbsp;Salwa El Gamal ,&nbsp;Basheer Youssef","doi":"10.1016/j.eij.2024.100571","DOIUrl":"10.1016/j.eij.2024.100571","url":null,"abstract":"<div><div>Vertical Broken Access Control (VBAC) vulnerability is one of the most commonly identified issues in web applications, posing significant risks. Consequently, addressing this pervasive threat is crucial for ensuring system confidentiality and integrity. Broken access control attack detector (BACAD) is a novel framework that leverages advanced AI techniques to neutralize VBAC exploits and attacks in real-time using a dynamic and practical technique. The detection process consists of two steps. The first step is user role classification using an advanced artificial intelligence (AI) model created in a learning phase. The learning phase includes BACAD initial configuration and application user roles traffic generation used for AI model training. The AI model at the core of BACAD analyzes web requests and responses utilizing a robust feature extraction, and dynamic hyperparameter tuning to ensure optimal performance across diverse scenarios. The second step is the decision step, which determines whether the incoming request–response pair is benign or an attack by validating it vs the BACAD session information set. The evaluation against a spectrum of real-world and demonstration web applications highlights remarkable efficiency in detecting VBAC exploits, providing robust application protection against different sets of VBAC attacks. Furthermore, it shows that BACAD addresses the VBAC problem by presenting an applicable, dynamic, flexible, and technology-independent solution to counter VBAC vulnerability risks. Thus, BACAD contributes significantly to the ongoing efforts aimed at enhancing web application security.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"28 ","pages":"Article 100571"},"PeriodicalIF":5.0,"publicationDate":"2024-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142658094","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}