Egyptian Informatics Journal最新文献_第6页

HD-MVCNN: High-density ECG signal based diabetic prediction and classification using multi-view convolutional neural network HD-MVCNN：利用多视角卷积神经网络进行基于高密度心电信号的糖尿病预测和分类

IF 5 3区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Egyptian Informatics Journal

Pub Date : 2024-11-06 DOI: 10.1016/j.eij.2024.100573

D. Santhakumar , K. Dhana Shree , M. Buvanesvari , A. Saran Kumar , Ayodeji Olalekan Salau

Diabetes mellitus, also known as diabetes, is a medical condition marked by high blood sugar levels and impacts a large population worldwide. Treating diabetes is not feasible. It can be managed. Hence, it is crucial to promptly identify a diagnosis of diabetes. This study explores the effects of diabetes on the heart, focusing on heart rate variability (HRV) signals, which can offer valuable information about the existence and seriousness of diabetes through the evaluation of diabetes-related heart problems. Extracting crucial data from the irregular and nonlinear HRV signal can be quite challenging. Studying cardiac diagnostics involves a thorough analysis of electrocardiogram (ECG) signals. Traditional electrocardiogram recordings utilize twelve channels, each capturing a complex combination of activities originating from different regions of the heart. Examining ECG signals recorded on the body’s surface may not be an effective method for studying and diagnosing diabetic issues. The study introduces a research proposal utilizing a high-density resolution electrocardiogram (ECG) system with a minimum of 64 channels and multi-view convolutional neural network classification (HD-MVCNN) to address the mentioned challenges. This framework may help identify the hypoglycaemia effects on brain regions, leading to decreased complexity and increased theta and delta power during scalp electrocardiogram procedures. The convolutional architectural model primarily contributes to enhancement and optimization through its Stochastic Gradient Descent (SGD) along with convolutional layers and according to results, the HD-MVCNN demonstrated better stability and accuracy in comparison to traditional classification models. Thus, HD-MVCNN shows promise as a powerful method for classifying features in diabetes clinical data.

糖尿病又称糖尿，是一种以高血糖为特征的病症，影响着全球众多人口。治疗糖尿病是不可行的。糖尿病是可以控制的。因此，及时确诊糖尿病至关重要。这项研究探讨了糖尿病对心脏的影响，重点是心率变异性（HRV）信号，它可以通过评估与糖尿病相关的心脏问题，提供有关糖尿病存在和严重程度的宝贵信息。从不规则和非线性心率变异信号中提取关键数据是一项相当具有挑战性的工作。研究心脏诊断需要对心电图（ECG）信号进行全面分析。传统的心电图记录使用十二个通道，每个通道捕捉来自心脏不同区域的复杂活动组合。检查体表记录的心电信号可能不是研究和诊断糖尿病问题的有效方法。本研究提出了一项研究建议，利用至少有 64 个通道的高密度分辨率心电图（ECG）系统和多视图卷积神经网络分类（HD-MVCNN）来应对上述挑战。该框架可帮助识别低血糖对大脑区域的影响，从而降低头皮心电图程序的复杂性并增加θ和δ功率。卷积架构模型主要通过其随机梯度下降（SGD）和卷积层来增强和优化，根据结果，与传统分类模型相比，HD-MVCNN 表现出更好的稳定性和准确性。因此，HD-MVCNN有望成为一种对糖尿病临床数据特征进行分类的强大方法。

{"title":"HD-MVCNN: High-density ECG signal based diabetic prediction and classification using multi-view convolutional neural network","authors":"D. Santhakumar , K. Dhana Shree , M. Buvanesvari , A. Saran Kumar , Ayodeji Olalekan Salau","doi":"10.1016/j.eij.2024.100573","DOIUrl":"10.1016/j.eij.2024.100573","url":null,"abstract":"<div><div>Diabetes mellitus, also known as diabetes, is a medical condition marked by high blood sugar levels and impacts a large population worldwide. Treating diabetes is not feasible. It can be managed. Hence, it is crucial to promptly identify a diagnosis of diabetes. This study explores the effects of diabetes on the heart, focusing on heart rate variability (HRV) signals, which can offer valuable information about the existence and seriousness of diabetes through the evaluation of diabetes-related heart problems. Extracting crucial data from the irregular and nonlinear HRV signal can be quite challenging. Studying cardiac diagnostics involves a thorough analysis of electrocardiogram (ECG) signals. Traditional electrocardiogram recordings utilize twelve channels, each capturing a complex combination of activities originating from different regions of the heart. Examining ECG signals recorded on the body’s surface may not be an effective method for studying and diagnosing diabetic issues. The study introduces a research proposal utilizing a high-density resolution electrocardiogram (ECG) system with a minimum of 64 channels and multi-view convolutional neural network classification (HD-MVCNN) to address the mentioned challenges. This framework may help identify the hypoglycaemia effects on brain regions, leading to decreased complexity and increased theta and delta power during scalp electrocardiogram procedures. The convolutional architectural model primarily contributes to enhancement and optimization through its Stochastic Gradient Descent (SGD) along with convolutional layers and according to results, the HD-MVCNN demonstrated better stability and accuracy in comparison to traditional classification models. Thus, HD-MVCNN shows promise as a powerful method for classifying features in diabetes clinical data.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"28 ","pages":"Article 100573"},"PeriodicalIF":5.0,"publicationDate":"2024-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142593126","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}

引用次数: 0

A hybrid encryption algorithm based approach for secure privacy protection of big data in hospitals 基于混合加密算法的医院大数据隐私安全保护方法

IF 5 3区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Egyptian Informatics Journal

Pub Date : 2024-11-05 DOI: 10.1016/j.eij.2024.100569

Wei Li , Qian Huang

Aiming at the hidden danger of information security caused by the lack of medical big data information security firewall, this paper proposes a security privacy protection method for hospital big data based on hybrid encryption algorithm. First, collect hospital big data including hospital medical business system, mobile wearable devices and big health data; Secondly, use byte changes to compress hospital big data to achieve safe transmission of hospital big data; Then, the hospital sender uses the AES session key to encrypt the hospital big data and the ECC public key to encrypt the AES session key, uses SHA-1 to calculate the hash value of the medical big data, and uses the ECC public key to sign the hash value; The hospital receiver uses the ECC private key to verify the signature, and decrypts the AES session key using the ECC private key. After the AES session key decrypts, the hospital big data, the hospital big data security privacy protection is completed. The experimental results show that the method is superior to conventional ECC algorithm or RSA and AES hybrid encryption algorithm in terms of encryption and decryption time and security strength. The average correlation coefficient of encrypted hospital big data is only 0.0576, and the RL curve value is low and gentle. The encrypted data has good scrambling effect and low privacy leakage probability, which ensures the confidentiality and integrity of medical data in the transmission process.

针对医疗大数据信息安全防火墙缺失带来的信息安全隐患，本文提出了一种基于混合加密算法的医院大数据安全隐私保护方法。首先，采集医院大数据，包括医院医疗业务系统、移动可穿戴设备、健康大数据等；其次，利用字节变化对医院大数据进行压缩，实现医院大数据的安全传输；然后，医院发送方使用 AES 会话密钥对医院大数据进行加密，使用 ECC 公钥对 AES 会话密钥进行加密，使用 SHA-1 计算医疗大数据的哈希值，使用 ECC 公钥对哈希值进行签名；医院接收方使用 ECC 私钥验证签名，使用 ECC 私钥对 AES 会话密钥进行解密。AES 会话密钥解密后，医院大数据的安全隐私保护就完成了。实验结果表明，该方法在加解密时间和安全强度方面均优于传统的ECC算法或RSA与AES混合加密算法。加密后的医院大数据平均相关系数仅为 0.0576，RL 曲线值较低且平缓。加密后的数据加扰效果好，隐私泄露概率低，保证了医疗数据在传输过程中的保密性和完整性。

{"title":"A hybrid encryption algorithm based approach for secure privacy protection of big data in hospitals","authors":"Wei Li , Qian Huang","doi":"10.1016/j.eij.2024.100569","DOIUrl":"10.1016/j.eij.2024.100569","url":null,"abstract":"<div><div>Aiming at the hidden danger of information security caused by the lack of medical big data information security firewall, this paper proposes a security privacy protection method for hospital big data based on hybrid encryption algorithm. First, collect hospital big data including hospital medical business system, mobile wearable devices and big health data; Secondly, use byte changes to compress hospital big data to achieve safe transmission of hospital big data; Then, the hospital sender uses the AES session key to encrypt the hospital big data and the ECC public key to encrypt the AES session key, uses SHA-1 to calculate the hash value of the medical big data, and uses the ECC public key to sign the hash value; The hospital receiver uses the ECC private key to verify the signature, and decrypts the AES session key using the ECC private key. After the AES session key decrypts, the hospital big data, the hospital big data security privacy protection is completed. The experimental results show that the method is superior to conventional ECC algorithm or RSA and AES hybrid encryption algorithm in terms of encryption and decryption time and security strength. The average correlation coefficient of encrypted hospital big data is only 0.0576, and the RL curve value is low and gentle. The encrypted data has good scrambling effect and low privacy leakage probability, which ensures the confidentiality and integrity of medical data in the transmission process.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"28 ","pages":"Article 100569"},"PeriodicalIF":5.0,"publicationDate":"2024-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142586849","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}

引用次数: 0

A new probabilistic linguistic decision-making process based on PL-BWM and improved three-way TODIM methods 基于 PL-BWM 和改进的三向 TODIM 方法的新概率语言决策过程

IF 5 3区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Egyptian Informatics Journal

Pub Date : 2024-11-04 DOI: 10.1016/j.eij.2024.100567

Jie Chen , Chuancun Yin

Probabilistic linguistic term sets (PLTSs) provide a flexible tool to express linguistic preferences, which allow decision-makers to label linguistic information with different probabilities. In this paper, a method based on a PLTS is proposed to address multi-criteria decision-making problems (MCDM). We develop the theory of PLTSs and put forward a novel best–worst method (BWM), termed PL-BWM, based on PLTS. Our method fully reflects the preference information of decision-makers and accurately provides the importance level of the criteria. The combined weight of the criteria is obtained by merging PL-BWM-based subjective weights and similarity minimization-based objective weights. Upon introducing a three-way decision system to improve the TODIM method, a novel three-way TODIM method is proposed and showcased on an optimal new energy vehicle selection problem. The effectiveness and accuracy of the proposed method are verified by sensitivity analysis and comparative analysis. Our approach paves the way for new developments in solving MCDM problems and for novel applications in otherwise difficult ranking problems.

概率语言术语集（PLTS）为表达语言偏好提供了一种灵活的工具，它允许决策者用不同的概率标记语言信息。本文提出了一种基于 PLTS 的方法来解决多标准决策问题（MCDM）。我们发展了 PLTS 的理论，并在 PLTS 的基础上提出了一种新颖的最佳-最差方法（BWM），称为 PL-BWM。我们的方法充分反映了决策者的偏好信息，并准确提供了标准的重要程度。通过合并基于 PL-BWM 的主观权重和基于相似性最小化的客观权重，可以得到标准的综合权重。在引入三向决策系统改进 TODIM 方法后，提出了一种新颖的三向 TODIM 方法，并在新能源汽车最优选择问题上进行了展示。通过灵敏度分析和对比分析，验证了所提方法的有效性和准确性。我们的方法为解决 MCDM 问题的新发展和其他困难排序问题的新应用铺平了道路。

{"title":"A new probabilistic linguistic decision-making process based on PL-BWM and improved three-way TODIM methods","authors":"Jie Chen , Chuancun Yin","doi":"10.1016/j.eij.2024.100567","DOIUrl":"10.1016/j.eij.2024.100567","url":null,"abstract":"<div><div>Probabilistic linguistic term sets (PLTSs) provide a flexible tool to express linguistic preferences, which allow decision-makers to label linguistic information with different probabilities. In this paper, a method based on a PLTS is proposed to address multi-criteria decision-making problems (MCDM). We develop the theory of PLTSs and put forward a novel best–worst method (BWM), termed PL-BWM, based on PLTS. Our method fully reflects the preference information of decision-makers and accurately provides the importance level of the criteria. The combined weight of the criteria is obtained by merging PL-BWM-based subjective weights and similarity minimization-based objective weights. Upon introducing a three-way decision system to improve the TODIM method, a novel three-way TODIM method is proposed and showcased on an optimal new energy vehicle selection problem. The effectiveness and accuracy of the proposed method are verified by sensitivity analysis and comparative analysis. Our approach paves the way for new developments in solving MCDM problems and for novel applications in otherwise difficult ranking problems.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"28 ","pages":"Article 100567"},"PeriodicalIF":5.0,"publicationDate":"2024-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142578487","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}

引用次数: 0

Interval valued inventory model with different payment strategies for green products under interval valued Grey Wolf optimizer Algorithm fitness function 区间值灰狼优化算法拟合函数下不同付款策略的绿色产品区间值库存模型

IF 5 3区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Egyptian Informatics Journal

Pub Date : 2024-11-02 DOI: 10.1016/j.eij.2024.100561

Subhash Chandra Das , Md. Al-Amin Khan , Ali Akbar Shaikh , Adel Fahad Alrasheedi

Numerous studies have explored pricing and lot-sizing strategies for various payment methods, but most have focused primarily on the buyer’s perspective. This study, however, approaches these strategies from a different perspective, incorporating key and relevant factors often overlooked. The volume of sales increases when a seller accepts a buyer’s credit. However, it reduces sales volume when a seller requests a buyer make a payment in advance. To boost sales and profitability, a vendor occasionally provides a price reduction in exchange for a down payment. Demanding a down payment from a customer earns interest and carries without any risk of default. When a vendor offers customers the option to pay with credit, a higher delay payment period facility plan may boost sales volume, but it also increases the risk of default. To maximize profit per unit of time, the vendor aims to simultaneously determine the optimal selling price, replenishment schedule, and payment method. This is achieved by comparing and calculating the vendor’s profit per time unit for credit, cash, and advance payment options. This is done by comparing and calculating the seller’s profit for each piece of time for credit, cash, and advance payments. The following managerial impacts are highlighted by means of numerical analyses: (1) A particular payment type, among the three available options, yields the seller’s highest profit under certain conditions. (2) It is vitally crucial for a vendor to provide a price reduction if an advance payment is required. (3) Advance payment results in higher profit than delayed payment if sales volume does not significantly fall while switching from credit to advance payments, or vice versa. To solve the optimization problem, a popular metaheuristic algorithm (viz., Grey Wolf Optimizer) is used and finally performed a post optimality analysis for making a fruitful conclusion.

许多研究探讨了各种支付方式的定价和批量大小策略，但大多数研究主要侧重于买方的角度。而本研究则从另一个角度探讨了这些策略，并纳入了经常被忽视的关键相关因素。当卖方接受买方赊账时，销售量会增加。然而，当卖方要求买方提前付款时，销售量就会减少。为了提高销售额和利润率，卖方偶尔会降价以换取首付款。要求客户支付首付款可以赚取利息，而且没有任何违约风险。当供应商向客户提供赊账付款选择时，较高的延迟付款期设施计划可能会提高销售量，但也会增加违约风险。为了使单位时间内的利润最大化，供应商要同时确定最佳销售价格、补货计划和付款方式。具体做法是比较并计算供应商在赊账、现金和预付款方式下的单位时间利润。具体做法是比较和计算赊账、现金和预付款方式下卖方在每个时间单位的利润。通过数值分析，突出了以下管理影响：（1）在特定条件下，三种可用付款方式中的某一种付款方式能为卖方带来最高利润。(2) 如果需要预付款，卖方必须降价。(3) 如果在从赊销转为预付款的过程中，销售量没有明显下降，则预付款的利润高于延迟付款，反之亦然。为了解决优化问题，我们使用了一种流行的元启发式算法（即灰狼优化器），最后进行了优化后分析，得出了富有成效的结论。

{"title":"Interval valued inventory model with different payment strategies for green products under interval valued Grey Wolf optimizer Algorithm fitness function","authors":"Subhash Chandra Das , Md. Al-Amin Khan , Ali Akbar Shaikh , Adel Fahad Alrasheedi","doi":"10.1016/j.eij.2024.100561","DOIUrl":"10.1016/j.eij.2024.100561","url":null,"abstract":"<div><div>Numerous studies have explored pricing and lot-sizing strategies for various payment methods, but most have focused primarily on the buyer’s perspective. This study, however, approaches these strategies from a different perspective, incorporating key and relevant factors often overlooked. The volume of sales increases when a seller accepts a buyer’s credit. However, it reduces sales volume when a seller requests a buyer make a payment in advance. To boost sales and profitability, a vendor occasionally provides a price reduction in exchange for a down payment. Demanding a down payment from a customer earns interest and carries without any risk of default. When a vendor offers customers the option to pay with credit, a higher delay payment period facility plan may boost sales volume, but it also increases the risk of default. To maximize profit per unit of time, the vendor aims to simultaneously determine the optimal selling price, replenishment schedule, and payment method. This is achieved by comparing and calculating the vendor’s profit per time unit for credit, cash, and advance payment options. This is done by comparing and calculating the seller’s profit for each piece of time for credit, cash, and advance payments. The following managerial impacts are highlighted by means of numerical analyses: (1) A particular payment type, among the three available options, yields the seller’s highest profit under certain conditions. (2) It is vitally crucial for a vendor to provide a price reduction if an advance payment is required. (3) Advance payment results in higher profit than delayed payment if sales volume does not significantly fall while switching from credit to advance payments, or vice versa. To solve the optimization problem, a popular metaheuristic algorithm (viz., Grey Wolf Optimizer) is used and finally performed a post optimality analysis for making a fruitful conclusion.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"28 ","pages":"Article 100561"},"PeriodicalIF":5.0,"publicationDate":"2024-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142571498","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}

引用次数: 0

Intelligent SDN to enhance security in IoT networks 智能 SDN 提高物联网网络的安全性

IF 5 3区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Egyptian Informatics Journal

Pub Date : 2024-10-30 DOI: 10.1016/j.eij.2024.100564

Safi Ibrahim , Aya M. Youssef , Mahmoud Shoman , Sanaa Taha

Software-defined networking (SDN) is a revolutionary technology that has revolutionised network management by providing flexibility and adaptability. As the popularity of SDN increases, it is crucial to address security vulnerabilities in these dynamic networks. This paper proposes a framework for enhancing security in SDN by utilising three separate Deep Learning models, namely Deep Neural Network (DNN), Convolutional Neural Network (CNN), and Long Short-Term Memory (LSTM). This framework is utilised for the InSDN dataset, a huge dataset specifically created for SDN security research. The dataset consists of a total of 343,939 instances, encompassing both normal and attack traffic. The regular data yields a sum of 68,424, whereas the attack traffic comprises 275,515 occurrences. This study employs multiclassification algorithms to precisely detect and categorise diverse security threats in SDN. The InSDN dataset faces issues related to class imbalance, which are addressed by using the Synthetic Minority Over-sampling Technique (SMOTE). The SMOTE technique is utilised to create artificial instances of the underrepresented class, hence achieving a more equitable distribution of security hazards within the dataset. This strategy improves the efficacy of multiclassification techniques, ultimately resulting in greater accuracy in the identification and classification of different security threats in SDN environments. The initial DNN model exhibited satisfactory performance, with an accuracy of 87%. The second CNN model demonstrated strong and consistent performance, with an accuracy rate of 99%. In addition, an LSTM model attained a 90% accuracy rate.

软件定义网络（SDN）是一项革命性技术，通过提供灵活性和适应性彻底改变了网络管理。随着 SDN 的普及，解决这些动态网络中的安全漏洞至关重要。本文提出了一个框架，通过利用三个独立的深度学习模型（即深度神经网络（DNN）、卷积神经网络（CNN）和长短期记忆（LSTM））来增强 SDN 的安全性。该框架用于 InSDN 数据集，这是一个专为 SDN 安全研究而创建的庞大数据集。该数据集由 343,939 个实例组成，包括正常流量和攻击流量。正常数据的总和为 68,424 次，而攻击流量包括 275,515 次。本研究采用了多分类算法来精确检测和分类 SDN 中的各种安全威胁。InSDN 数据集面临着与类不平衡相关的问题，通过使用合成少数群体过度采样技术（SMOTE）解决了这些问题。SMOTE 技术用于创建代表性不足类别的人工实例，从而在数据集中实现更公平的安全隐患分布。这一策略提高了多分类技术的功效，最终提高了在 SDN 环境中识别和分类不同安全威胁的准确性。最初的 DNN 模型表现令人满意，准确率达到 87%。第二个 CNN 模型表现出强大而稳定的性能，准确率达到 99%。此外，LSTM 模型的准确率也达到了 90%。

{"title":"Intelligent SDN to enhance security in IoT networks","authors":"Safi Ibrahim , Aya M. Youssef , Mahmoud Shoman , Sanaa Taha","doi":"10.1016/j.eij.2024.100564","DOIUrl":"10.1016/j.eij.2024.100564","url":null,"abstract":"<div><div>Software-defined networking (SDN) is a revolutionary technology that has revolutionised network management by providing flexibility and adaptability. As the popularity of SDN increases, it is crucial to address security vulnerabilities in these dynamic networks. This paper proposes a framework for enhancing security in SDN by utilising three separate Deep Learning models, namely Deep Neural Network (DNN), Convolutional Neural Network (CNN), and Long Short-Term Memory (LSTM). This framework is utilised for the InSDN dataset, a huge dataset specifically created for SDN security research. The dataset consists of a total of 343,939 instances, encompassing both normal and attack traffic. The regular data yields a sum of 68,424, whereas the attack traffic comprises 275,515 occurrences. This study employs multiclassification algorithms to precisely detect and categorise diverse security threats in SDN. The InSDN dataset faces issues related to class imbalance, which are addressed by using the Synthetic Minority Over-sampling Technique (SMOTE). The SMOTE technique is utilised to create artificial instances of the underrepresented class, hence achieving a more equitable distribution of security hazards within the dataset. This strategy improves the efficacy of multiclassification techniques, ultimately resulting in greater accuracy in the identification and classification of different security threats in SDN environments. The initial DNN model exhibited satisfactory performance, with an accuracy of 87%. The second CNN model demonstrated strong and consistent performance, with an accuracy rate of 99%. In addition, an LSTM model attained a 90% accuracy rate.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"28 ","pages":"Article 100564"},"PeriodicalIF":5.0,"publicationDate":"2024-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142553747","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}

引用次数: 0

Fuzzy operator infrared image deblurring algorithm for image blurring in dragon boat races 用于龙舟赛图像模糊处理的模糊算子红外图像去模糊算法

IF 5 3区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Egyptian Informatics Journal

Pub Date : 2024-10-28 DOI: 10.1016/j.eij.2024.100568

Xiao Tang , Yuan Shen , Genwei Zhu

To address the issues of poor robustness and weak generalization in existing infrared image deblurring methods, a fuzzy operator-based algorithm is proposed to solve the fuzzy imaging in dragon boat races. The experiment showed that the models trained utilizing original and synthesized datasets had very small differences in peak signal-to-noise ratio and structural similarity performance indicators, and the evaluation results were close. For a blurry image with 19 pixels, the number of blurry pixels extracted by the research algorithm was 22, with a difference of 3 pixels. For a blurry image with 35 pixels, the algorithm extracted 34 blurry pixels, with a difference of 1 pixel. This indicated that the deblurring result of the algorithm was accurate. In terms of peak signal-to-noise ratio and structural similarity, the peak signal-to-noise ratio and structure similarity were 30.98 dB and 0.921, respectively, both of which were the optimal values in all algorithms. In terms of the change of pixel gray value, the simulated blur length of the research method was 19 pixels, and the actual blur length was 20 pixels far less than 30 pixels. The results verified the effectiveness and significance of the algorithm for deblurring of dragon boat competition infrared images.

针对现有红外图像去模糊方法鲁棒性差、泛化能力弱的问题，提出了一种基于模糊算子的算法来解决龙舟赛中的模糊成像问题。实验表明，利用原始数据集和合成数据集训练的模型在峰值信噪比和结构相似性性能指标上差异很小，评价结果接近。对于 19 像素的模糊图像，研究算法提取的模糊像素数为 22，相差 3 像素。对于 35 像素的模糊图像，该算法提取了 34 个模糊像素，相差 1 个像素。这表明算法的去模糊结果是准确的。在峰值信噪比和结构相似度方面，峰值信噪比和结构相似度分别为 30.98 dB 和 0.921，均为所有算法的最优值。在像素灰度值变化方面，研究方法的模拟模糊长度为 19 像素，实际模糊长度为 20 像素，远小于 30 像素。结果验证了该算法在龙舟比赛红外图像去模糊方面的有效性和意义。

{"title":"Fuzzy operator infrared image deblurring algorithm for image blurring in dragon boat races","authors":"Xiao Tang , Yuan Shen , Genwei Zhu","doi":"10.1016/j.eij.2024.100568","DOIUrl":"10.1016/j.eij.2024.100568","url":null,"abstract":"<div><div>To address the issues of poor robustness and weak generalization in existing infrared image deblurring methods, a fuzzy operator-based algorithm is proposed to solve the fuzzy imaging in dragon boat races. The experiment showed that the models trained utilizing original and synthesized datasets had very small differences in peak signal-to-noise ratio and structural similarity performance indicators, and the evaluation results were close. For a blurry image with 19 pixels, the number of blurry pixels extracted by the research algorithm was 22, with a difference of 3 pixels. For a blurry image with 35 pixels, the algorithm extracted 34 blurry pixels, with a difference of 1 pixel. This indicated that the deblurring result of the algorithm was accurate. In terms of peak signal-to-noise ratio and structural similarity, the peak signal-to-noise ratio and structure similarity were 30.98 dB and 0.921, respectively, both of which were the optimal values in all algorithms. In terms of the change of pixel gray value, the simulated blur length of the research method was 19 pixels, and the actual blur length was 20 pixels far less than 30 pixels. The results verified the effectiveness and significance of the algorithm for deblurring of dragon boat competition infrared images.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"28 ","pages":"Article 100568"},"PeriodicalIF":5.0,"publicationDate":"2024-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142530564","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}

引用次数: 0

A Lightweight malware detection technique based on hybrid fuzzy simulated annealing clustering in Android apps 基于混合模糊模拟退火聚类的安卓应用程序轻量级恶意软件检测技术

IF 5 3区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Egyptian Informatics Journal

Pub Date : 2024-10-19 DOI: 10.1016/j.eij.2024.100560

Collins Chimeleze , Norziana Jamil , Nazik Alturki , Zuhaira Muhammad Zain

The growing complexity of cyber threats has shifted the focus from merely identifying threats to detecting their origins, resulting in stronger defenses against malware. Traditional detection techniques are often inadequate against increasingly sophisticated malware, prompting this research article to propose a new clustering method—fuzzy C-mean simulated annealing (FCMSA)—to enhance malware detection through machine learning. The FCMSA clustering technique improves performance by minimizing vulnerabilities, reducing outliers, and optimizing large datasets. The proposed technique selects high-quality clusters from Android app permissions and, using lightGBM, classifies Android malware. Experimental results show that the proposed FCMSA-GBM technique achieves superior accuracy (99.21%) and precision (99.70%) compared to other prevalent cluster-based Android malware detection techniques, while also lowering error rates and execution time.

网络威胁日益复杂，人们已将关注点从单纯识别威胁转移到检测威胁来源，从而加强对恶意软件的防御。传统的检测技术往往不足以应对日益复杂的恶意软件，因此本研究文章提出了一种新的聚类方法--模糊C均值模拟退火（FCMSA）--通过机器学习来增强恶意软件的检测能力。FCMSA 聚类技术通过最小化漏洞、减少异常值和优化大型数据集来提高性能。该技术从安卓应用程序权限中选择高质量聚类，并使用 lightGBM 对安卓恶意软件进行分类。实验结果表明，与其他流行的基于聚类的安卓恶意软件检测技术相比，所提出的 FCMSA-GBM 技术实现了更高的准确率（99.21%）和精确度（99.70%），同时还降低了错误率和执行时间。

引用次数: 0

Deep Learning-Assisted Compound Bioactivity Estimation Framework 深度学习辅助化合物生物活性估算框架

IF 5 3区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Egyptian Informatics Journal

Pub Date : 2024-10-15 DOI: 10.1016/j.eij.2024.100558

Yasmine Eid Mahmoud Yousef , Ayman El-Kilany , Farid Ali , Yassin M. Nissan , Ehab E. Hassanein

Drug Discovery is a highly complicated process. On average, it takes six to twelve years to manufacture a new drug and have the product released in the market. It is of utmost importance to find methods that would accelerate the manufacturing process. This significant challenge in drug development can be addressed using deep learning techniques. The aim of this paper is to propose a deep learning-based framework that can help chemists examine compound biological activity in a more accurate manner. The proposed framework employs autoencoder for data representation of the compounds data, which is then classified using deep neural network followed by building a customized deep regression model to estimate an accurate value of the compound bioactivity. The proposed framework achieved an accuracy of 89% in autoencoder reconstruction error, 79.01% in classification, and MAE of 2.4 while predicting compound bioactivity using deep regression model.

药物研发是一个非常复杂的过程。生产一种新药并将产品投放市场平均需要 6 到 12 年的时间。因此，找到能加快生产过程的方法至关重要。深度学习技术可以解决药物研发中的这一重大挑战。本文旨在提出一种基于深度学习的框架，帮助化学家更准确地研究化合物的生物活性。所提出的框架采用自动编码器对化合物数据进行数据表示，然后使用深度神经网络对其进行分类，接着建立一个定制的深度回归模型来估算化合物生物活性的准确值。在使用深度回归模型预测化合物生物活性时，拟议框架的自动编码器重构误差准确率达到 89%，分类准确率达到 79.01%，MAE 为 2.4。

引用次数: 0

Delay probability in adaptive systems based on activation function of classical neural networks 基于经典神经网络激活函数的自适应系统中的延迟概率

IF 5 3区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Egyptian Informatics Journal

Pub Date : 2024-10-14 DOI: 10.1016/j.eij.2024.100555

Maja Lutovac Banduka , Vladimir Mladenović , Danijela Milosević , Vladimir Orlić , Asutosh Kar

Many improved algorithms have been proposed for nonlinear system designs. There is no single procedure for providing an algorithm with closed-form system response relations as a function of system parameters. In this paper, we illustrate a unique method for discrete-time digital nonlinear systems. Provides better insight into the analyzed system, algorithm, and processes. The main contribution is closed-form symbolic responses in the time domain and modifications of the implemented algorithm. A comparison of adaptive systems and neural networks is also presented. The design and analysis of nonlinear systems are more clearly simplified for either engineers or researchers without deep mathematical knowledge.

针对非线性系统设计提出了许多改进算法。目前还没有一种算法能提供与系统参数有关的闭式系统响应关系。在本文中，我们阐述了一种适用于离散时间数字非线性系统的独特方法。它能让我们更好地了解所分析的系统、算法和过程。主要贡献在于时域的闭式符号响应和对已实施算法的修改。此外，还对自适应系统和神经网络进行了比较。对于没有深厚数学知识的工程师或研究人员来说，非线性系统的设计和分析得到了更清晰的简化。

引用次数: 0

Heart disease prediction using autoencoder and DenseNet architecture 利用自动编码器和 DenseNet 架构预测心脏病

IF 5 3区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Egyptian Informatics Journal

Pub Date : 2024-10-14 DOI: 10.1016/j.eij.2024.100559

Norah Saleh Alghamdi , Mohammed Zakariah , Achyut Shankar , Wattana Viriyasitavat

Heart disease continues to be a prominent cause of death globally, emphasizing the critical requirement for precise prediction techniques and prompt therapies. This research presents a new method that utilizes the collective capabilities of autoencoder and DenseNet architectures to predict heart illness. Our study is based on the Heart Disease UCI Cleveland dataset, which includes 13 variables that cover clinical and demographic parameters such as age, sex, cholesterol levels, and exercise-induced angina. The dataset presents issues due to its varied attribute types, including category and numerical variables. Furthermore, our approach tackles these difficulties by utilizing a dense autoencoder model, which produced exceptional outcomes. The Model attained a mean accuracy of 99.67% on the Heart Disease UCI Cleveland dataset. Further testing showed it was resilient, with a test accuracy of 99.99%. In addition, the Model demonstrated outstanding macro precision, macro recall, and macro F1 score, with percentages of 99.98%, 99.97%, and 99.96%, respectively. In addition, our results indicate that combining autoencoder and DenseNet designs shows potential for predicting cardiac disease, with substantial enhancements in accuracy and performance metrics compared to current approaches. This methodology can improve clinical decision-making and patient outcomes in cardiovascular care by accurately finding and defining complex patterns within the data. Notwithstanding these encouraging outcomes, our investigation has constraints. The specific attributes of the dataset utilized may limit the applicability of our findings. Subsequent studies could examine the suitability of our method for various datasets and analyze supplementary variables that may improve forecast precision. Furthermore, it is necessary to conduct prospective validation studies to evaluate our strategy’s practical effectiveness in clinical environments.

心脏病仍然是全球死亡的一个主要原因，这就强调了对精确预测技术和及时治疗的迫切需求。本研究提出了一种新方法，利用自动编码器和 DenseNet 架构的集体能力来预测心脏病。我们的研究基于心脏病 UCI 克利夫兰数据集，该数据集包含 13 个变量，涵盖临床和人口统计参数，如年龄、性别、胆固醇水平和运动诱发的心绞痛。由于该数据集的属性类型多种多样，包括类别变量和数字变量，因此存在一些问题。此外，我们的方法利用密集的自动编码器模型解决了这些难题，并取得了卓越的成果。该模型在心脏病 UCI 克利夫兰数据集上的平均准确率达到 99.67%。进一步测试表明，该模型具有很强的适应能力，测试准确率达到 99.99%。此外，该模型的宏观精确度、宏观召回率和宏观 F1 分数也表现出色，分别达到 99.98%、99.97% 和 99.96%。此外，我们的研究结果表明，结合自动编码器和 DenseNet 设计具有预测心脏疾病的潜力，与目前的方法相比，准确率和性能指标都有大幅提高。这种方法能准确发现和定义数据中的复杂模式，从而改善心血管护理的临床决策和患者预后。尽管取得了这些令人鼓舞的成果，但我们的研究也有局限性。所使用数据集的特定属性可能会限制我们研究结果的适用性。后续研究可以检查我们的方法是否适用于各种数据集，并分析可能提高预测精度的补充变量。此外，有必要进行前瞻性验证研究，以评估我们的策略在临床环境中的实际效果。

{"title":"Heart disease prediction using autoencoder and DenseNet architecture","authors":"Norah Saleh Alghamdi , Mohammed Zakariah , Achyut Shankar , Wattana Viriyasitavat","doi":"10.1016/j.eij.2024.100559","DOIUrl":"10.1016/j.eij.2024.100559","url":null,"abstract":"<div><div>Heart disease continues to be a prominent cause of death globally, emphasizing the critical requirement for precise prediction techniques and prompt therapies. This research presents a new method that utilizes the collective capabilities of autoencoder and DenseNet architectures to predict heart illness. Our study is based on the Heart Disease UCI Cleveland dataset, which includes 13 variables that cover clinical and demographic parameters such as age, sex, cholesterol levels, and exercise-induced angina. The dataset presents issues due to its varied attribute types, including category and numerical variables. Furthermore, our approach tackles these difficulties by utilizing a dense autoencoder model, which produced exceptional outcomes. The Model attained a mean accuracy of 99.67% on the Heart Disease UCI Cleveland dataset. Further testing showed it was resilient, with a test accuracy of 99.99%. In addition, the Model demonstrated outstanding macro precision, macro recall, and macro F1 score, with percentages of 99.98%, 99.97%, and 99.96%, respectively. In addition, our results indicate that combining autoencoder and DenseNet designs shows potential for predicting cardiac disease, with substantial enhancements in accuracy and performance metrics compared to current approaches. This methodology can improve clinical decision-making and patient outcomes in cardiovascular care by accurately finding and defining complex patterns within the data. Notwithstanding these encouraging outcomes, our investigation has constraints. The specific attributes of the dataset utilized may limit the applicability of our findings. Subsequent studies could examine the suitability of our method for various datasets and analyze supplementary variables that may improve forecast precision. Furthermore, it is necessary to conduct prospective validation studies to evaluate our strategy’s practical effectiveness in clinical environments.</div></div>","PeriodicalId":56010,"journal":{"name":"Egyptian Informatics Journal","volume":"28 ","pages":"Article 100559"},"PeriodicalIF":5.0,"publicationDate":"2024-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142432234","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}

引用次数: 0