Pub Date : 2023-07-01DOI: 10.1109/CSCloud-EdgeCom58631.2023.00078
Hao Wu, Jun Qi, Yong Yue
As the second most common neurodegenerative disease in the world, Parkinson’s disease continues to affect the normal and healthy life of patients. In recent years, considerable progress has been made in studying the EEG of patients with Parkinson’s disease. Many EEG data of patients with Parkinson’s disease can be published, and more filtering algorithms and classification models suitable for EEG signals of Parkinson’s disease have been proposed. However, studying channel redundancy of EEG signals in Parkinson’s disease still faces challenges. The pathogenesis of Parkinson’s disease is still uncertain in medicine, and it is difficult to propose a channel selection scheme suitable for all patients with Parkinson’s disease. In this paper, the open UNM data set is used to extract multi-scale features based on the fourth-order Butterworth IIR filter and Wavelet Packet Transform. The channel selection is carried out by using single-channel verification. 12 and 25 channels with the relative best R2 scores were selected for the feature data set generated based on these two methods. The classification performance of data sets with and without channel selection was compared between the open and closed-eye data sets. The negative effect of open eye status on EEG classification of Parkinson’s disease was found, and the channel selection was used to improve the AUC by 1% in the same data set. Results showed that the proposed channel selection scheme can alleviate the overfitting phenomenon that occurred in the training set in the testing set while maintaining the classification accuracy.
{"title":"Machine Learning-based EEG Signal Classification of Parkinson’s Disease","authors":"Hao Wu, Jun Qi, Yong Yue","doi":"10.1109/CSCloud-EdgeCom58631.2023.00078","DOIUrl":"https://doi.org/10.1109/CSCloud-EdgeCom58631.2023.00078","url":null,"abstract":"As the second most common neurodegenerative disease in the world, Parkinson’s disease continues to affect the normal and healthy life of patients. In recent years, considerable progress has been made in studying the EEG of patients with Parkinson’s disease. Many EEG data of patients with Parkinson’s disease can be published, and more filtering algorithms and classification models suitable for EEG signals of Parkinson’s disease have been proposed. However, studying channel redundancy of EEG signals in Parkinson’s disease still faces challenges. The pathogenesis of Parkinson’s disease is still uncertain in medicine, and it is difficult to propose a channel selection scheme suitable for all patients with Parkinson’s disease. In this paper, the open UNM data set is used to extract multi-scale features based on the fourth-order Butterworth IIR filter and Wavelet Packet Transform. The channel selection is carried out by using single-channel verification. 12 and 25 channels with the relative best R2 scores were selected for the feature data set generated based on these two methods. The classification performance of data sets with and without channel selection was compared between the open and closed-eye data sets. The negative effect of open eye status on EEG classification of Parkinson’s disease was found, and the channel selection was used to improve the AUC by 1% in the same data set. Results showed that the proposed channel selection scheme can alleviate the overfitting phenomenon that occurred in the training set in the testing set while maintaining the classification accuracy.","PeriodicalId":56007,"journal":{"name":"Journal of Cloud Computing-Advances Systems and Applications","volume":"3 1","pages":"423-428"},"PeriodicalIF":4.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89681333","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}
Pub Date : 2023-07-01DOI: 10.1109/CSCloud-EdgeCom58631.2023.00024
Yan Li, Xin Su, Xin Liu, He Yi Mu, Y. Zheng, Shuping Wang
To better carry out early warning and control work for high-risk individuals in society, this paper proposes a risk assessment model based on graph attention networks. The model analyzes relevant background and relationship information of these individuals and constructs a knowledge graph accordingly. An improved graph attention mechanism is introduced to establish the risk assessment model. Real police character data was used to train and test the model, and experimental results indicated a prediction accuracy of 89.4%, with both accuracy and recall rates around 90%. This model can provide decision-making basis and technical support for early warning of public security personnel by identifying potential risks of high-risk individuals.
{"title":"Research on Risk Assessment Model for Social High-Risk Individuals Based on Graph Attention Network","authors":"Yan Li, Xin Su, Xin Liu, He Yi Mu, Y. Zheng, Shuping Wang","doi":"10.1109/CSCloud-EdgeCom58631.2023.00024","DOIUrl":"https://doi.org/10.1109/CSCloud-EdgeCom58631.2023.00024","url":null,"abstract":"To better carry out early warning and control work for high-risk individuals in society, this paper proposes a risk assessment model based on graph attention networks. The model analyzes relevant background and relationship information of these individuals and constructs a knowledge graph accordingly. An improved graph attention mechanism is introduced to establish the risk assessment model. Real police character data was used to train and test the model, and experimental results indicated a prediction accuracy of 89.4%, with both accuracy and recall rates around 90%. This model can provide decision-making basis and technical support for early warning of public security personnel by identifying potential risks of high-risk individuals.","PeriodicalId":56007,"journal":{"name":"Journal of Cloud Computing-Advances Systems and Applications","volume":"4 1","pages":"91-95"},"PeriodicalIF":4.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90070753","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}
Pub Date : 2023-07-01DOI: 10.1109/CSCloud-EdgeCom58631.2023.00074
Lijun Xiao, Dezhi Han, Sisi Zhou, Nengxiang Xu, Lin Chen, Siqi Xie
In recent years, data privacy security has been widely and highly valued by countries around the world. In the context of European Union’s General Data Protection Regulation (GDPR), the regulatory requirements of laws and regulations are becoming increasingly strict, bringing huge impacts and challenges to enterprises with user’s personal data such as internet services and financial technology. Up to a point, federal learning ensures data privacy by storing and processing personal data locally. However, due to malicious clients or central servers being able to launch attacks on global models or user privacy data, the security of federated learning is questioned, introducing blockchain into the federated learning framework is a feasible solution to address these data security issues. In this work, the concept of Blockchain (BC), Federated Learning (FL), GDPR and other similar data protection laws are presented, where a Blockchain-empowered Federated Learning (BC-empowered FL) framework is introduced. The challenges on complying with the GDPR are described, and the solutions or principles for improving the GDPR-compliance of BC-empowered FL systems are analyzed, sorting out the differences and connections among the GDPR-compliance methods yet laying a foundation to design legal and compliant applications for different domains and scenarios which need touch upon the user’s personal data.
{"title":"A Blockchain-empowered Federated Learning Framework Supprting GDPR-compliance","authors":"Lijun Xiao, Dezhi Han, Sisi Zhou, Nengxiang Xu, Lin Chen, Siqi Xie","doi":"10.1109/CSCloud-EdgeCom58631.2023.00074","DOIUrl":"https://doi.org/10.1109/CSCloud-EdgeCom58631.2023.00074","url":null,"abstract":"In recent years, data privacy security has been widely and highly valued by countries around the world. In the context of European Union’s General Data Protection Regulation (GDPR), the regulatory requirements of laws and regulations are becoming increasingly strict, bringing huge impacts and challenges to enterprises with user’s personal data such as internet services and financial technology. Up to a point, federal learning ensures data privacy by storing and processing personal data locally. However, due to malicious clients or central servers being able to launch attacks on global models or user privacy data, the security of federated learning is questioned, introducing blockchain into the federated learning framework is a feasible solution to address these data security issues. In this work, the concept of Blockchain (BC), Federated Learning (FL), GDPR and other similar data protection laws are presented, where a Blockchain-empowered Federated Learning (BC-empowered FL) framework is introduced. The challenges on complying with the GDPR are described, and the solutions or principles for improving the GDPR-compliance of BC-empowered FL systems are analyzed, sorting out the differences and connections among the GDPR-compliance methods yet laying a foundation to design legal and compliant applications for different domains and scenarios which need touch upon the user’s personal data.","PeriodicalId":56007,"journal":{"name":"Journal of Cloud Computing-Advances Systems and Applications","volume":"40 1","pages":"399-404"},"PeriodicalIF":4.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79663109","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}
Pub Date : 2023-07-01DOI: 10.1109/CSCloud-EdgeCom58631.2023.00020
Qinyang Chen, Keming Wang, Tao Xie
Vehicular fog computing (VFC) has emerged as a popular paradigm in the Internet of Vehicles (IoV) by replacing cloud servers with edge fog servers, thereby reducing communication latency and improving efficiency. However, data collection in VFC poses significant security challenges, particularly with respect to the privacy of location data, which is essential for effective vehicle data collection. While some traditional location privacy protection schemes have been developed, they fail to address the availability of location data and the traceability of special situations. To overcome these limitations, we propose a traceable vehicle location privacy protection scheme in VFC that ensures the availability of location data by using fuzzy location information and achieves the traceability of location data through a secret sharing scheme. The simulation results demonstrate its effectiveness and feasibility.
{"title":"A Traceable Location Privacy Preserving Scheme for Data Collection in Vehicular Fog Computing","authors":"Qinyang Chen, Keming Wang, Tao Xie","doi":"10.1109/CSCloud-EdgeCom58631.2023.00020","DOIUrl":"https://doi.org/10.1109/CSCloud-EdgeCom58631.2023.00020","url":null,"abstract":"Vehicular fog computing (VFC) has emerged as a popular paradigm in the Internet of Vehicles (IoV) by replacing cloud servers with edge fog servers, thereby reducing communication latency and improving efficiency. However, data collection in VFC poses significant security challenges, particularly with respect to the privacy of location data, which is essential for effective vehicle data collection. While some traditional location privacy protection schemes have been developed, they fail to address the availability of location data and the traceability of special situations. To overcome these limitations, we propose a traceable vehicle location privacy protection scheme in VFC that ensures the availability of location data by using fuzzy location information and achieves the traceability of location data through a secret sharing scheme. The simulation results demonstrate its effectiveness and feasibility.","PeriodicalId":56007,"journal":{"name":"Journal of Cloud Computing-Advances Systems and Applications","volume":"2 1","pages":"65-71"},"PeriodicalIF":4.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79945372","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}
Pub Date : 2023-07-01DOI: 10.1109/CSCloud-EdgeCom58631.2023.00038
Chenying Xu, Yanfei Yin, Yingwen Chen
With the development of cloud computing, data owners generally use cloud services to reduce storage and computing overhead. However, data stored in cloud servers is out of the direct control of the data owners, causing serious security issues. Access control mechanism based on cryptography can effectively protect the security of cloud data and prevent unauthorized access to it. Nevertheless, users may redistribute data to other users after obtaining it, which can harm the copyright interests of the data owner. To address this issue, this paper proposes a secret and traceable approach for cloud data sharing. We combine the lattice-based proxy re-encryption with digital watermarking technology for redistribution tracking in cloud data sharing scenario. The lattice cipher used in this scheme is an encryption algorithm with homomorphic property based on the Ring-LWE problem. It has the advantage of reducing the ciphertext length and improving the efficiency of encryption. This paper demonstrates the feasibility and effectiveness of this scheme from a series of experiments. The encryption of an image of size 1024*1024 pixels takes only 9.78 seconds on average. The computational performance of this scheme is roughly 91.434% better than the existing data sharing schemes supporting redistribution tracking.
{"title":"A Secret and Traceable Approach for Cloud Data Sharing","authors":"Chenying Xu, Yanfei Yin, Yingwen Chen","doi":"10.1109/CSCloud-EdgeCom58631.2023.00038","DOIUrl":"https://doi.org/10.1109/CSCloud-EdgeCom58631.2023.00038","url":null,"abstract":"With the development of cloud computing, data owners generally use cloud services to reduce storage and computing overhead. However, data stored in cloud servers is out of the direct control of the data owners, causing serious security issues. Access control mechanism based on cryptography can effectively protect the security of cloud data and prevent unauthorized access to it. Nevertheless, users may redistribute data to other users after obtaining it, which can harm the copyright interests of the data owner. To address this issue, this paper proposes a secret and traceable approach for cloud data sharing. We combine the lattice-based proxy re-encryption with digital watermarking technology for redistribution tracking in cloud data sharing scenario. The lattice cipher used in this scheme is an encryption algorithm with homomorphic property based on the Ring-LWE problem. It has the advantage of reducing the ciphertext length and improving the efficiency of encryption. This paper demonstrates the feasibility and effectiveness of this scheme from a series of experiments. The encryption of an image of size 1024*1024 pixels takes only 9.78 seconds on average. The computational performance of this scheme is roughly 91.434% better than the existing data sharing schemes supporting redistribution tracking.","PeriodicalId":56007,"journal":{"name":"Journal of Cloud Computing-Advances Systems and Applications","volume":"19 1","pages":"173-180"},"PeriodicalIF":4.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77990304","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}
With the growing demand for communication and the rapid development of global mobile communication technology, the fourth generation of mobile communication can no longer meet the needs of people in their working lives for mobile network communication, and the fifth generation of mobile communication is born. This paper first introduces three strategies for the use of the 5G spectrum, namely spectrum dedication, spectrum re-farming, and spectrum sharing. Secondly, it introduces the three main application scenarios for the 5G spectrum, namely enhanced mobile broadband (eMBB), ultrahigh reliability and ultra-low latency services (URLLC), and massive IoT communications (MTC). The presentation concludes with a summary of the impact of the 5G spectrum in today’s society and an outlook on its future development.
{"title":"5G Spectrum Research","authors":"Peiyuan Zhu, Lijun Xiao, Shu Tan, Jiahong Cai, Yingzi Huo, Ronglin Zhang","doi":"10.1109/CSCloud-EdgeCom58631.2023.00035","DOIUrl":"https://doi.org/10.1109/CSCloud-EdgeCom58631.2023.00035","url":null,"abstract":"With the growing demand for communication and the rapid development of global mobile communication technology, the fourth generation of mobile communication can no longer meet the needs of people in their working lives for mobile network communication, and the fifth generation of mobile communication is born. This paper first introduces three strategies for the use of the 5G spectrum, namely spectrum dedication, spectrum re-farming, and spectrum sharing. Secondly, it introduces the three main application scenarios for the 5G spectrum, namely enhanced mobile broadband (eMBB), ultrahigh reliability and ultra-low latency services (URLLC), and massive IoT communications (MTC). The presentation concludes with a summary of the impact of the 5G spectrum in today’s society and an outlook on its future development.","PeriodicalId":56007,"journal":{"name":"Journal of Cloud Computing-Advances Systems and Applications","volume":"1 1","pages":"155-160"},"PeriodicalIF":4.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75111540","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}
Pub Date : 2023-07-01DOI: 10.1109/CSCloud-EdgeCom58631.2023.00043
Souradip Roy, Juan Li, Yan Bai
As the Internet of Things (IoT) becomes more prevalent, the need for intrusion detection systems (IDS) to protect against cyberattacks increases. However, the limited computing capabilities of IoT devices often require sending data to a centralized cloud for analysis, which can cause energy consumption, privacy issues, and data leakage. To address these problems, we propose a Federated Learning-based IDS that distributes learning to local devices without sending data to a centralized cloud. We also create lightweight local learners to accommodate IoT device limitations and locally adapted models to handle non-independent intrusion data distribution. We evaluate our method using NBaIoT and CICIDS-2017 datasets, and our results demonstrate comparable performance to centralized learning on metrics including accuracy, precision, and recall, while addressing privacy and data leakage concerns.
{"title":"Federated Learning-Based Intrusion Detection System for IoT Environments with Locally Adapted Model","authors":"Souradip Roy, Juan Li, Yan Bai","doi":"10.1109/CSCloud-EdgeCom58631.2023.00043","DOIUrl":"https://doi.org/10.1109/CSCloud-EdgeCom58631.2023.00043","url":null,"abstract":"As the Internet of Things (IoT) becomes more prevalent, the need for intrusion detection systems (IDS) to protect against cyberattacks increases. However, the limited computing capabilities of IoT devices often require sending data to a centralized cloud for analysis, which can cause energy consumption, privacy issues, and data leakage. To address these problems, we propose a Federated Learning-based IDS that distributes learning to local devices without sending data to a centralized cloud. We also create lightweight local learners to accommodate IoT device limitations and locally adapted models to handle non-independent intrusion data distribution. We evaluate our method using NBaIoT and CICIDS-2017 datasets, and our results demonstrate comparable performance to centralized learning on metrics including accuracy, precision, and recall, while addressing privacy and data leakage concerns.","PeriodicalId":56007,"journal":{"name":"Journal of Cloud Computing-Advances Systems and Applications","volume":"10 1","pages":"203-209"},"PeriodicalIF":4.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75258619","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}
Conducting binary function naming helps reverse engineers understand the internal workings of the code and perform malicious code analysis without accessing the source code. However, the loss of debugging information poses the challenge of insufficient high-level semantic information description for stripping binary code function naming. Meanwhile, the existing binary function naming scheme has one function label for only one sample. The long-tail effect of function labels for a single sample makes the machine learning-based prediction models face the challenge. To obtain a function correlation label and improve the propensity score of uncommon tail labels, we propose a multi-label learning-based binary function naming model BContext2Name. This model automatically generates relevant labels for binary function naming by function context information with the help of PfastreXML model. The experimental results show that BContext2Name can enrich function labels and alleviate the long-tail effect that exists for a single sample class. To obtain high-level semantics of binary functions, we align pseudocode and basic blocks based on disassembly and decompilation, identify concrete or abstract values of API parameters by variable tracking, and construct API-enhanced control flow graphs. Finally, a seq2seq neural network translation model with attention mechanism is constructed between function multi-label learning and enhanced control flow graphs. Experiments on the dataset reveal that the F1 values of the BContext2Name model improve by 3.55% and 15.23% over the state-of-the-art XFL and Nero, respectively. This indicates that function multi-label learning can provide accurate labels for binary functions and can help reverse analysts understand the inner working mechanism of binary code. Code and data for this evaluation are available at https://github.com/CSecurityZhongYuan/BContext2Name.
{"title":"BContext2Name: Naming Functions in Stripped Binaries with Multi-Label Learning and Neural Networks","authors":"Bing Xia, Yunxiang Ge, Ruinan Yang, Jiabin Yin, Jianmin Pang, Chongjun Tang","doi":"10.1109/CSCloud-EdgeCom58631.2023.00037","DOIUrl":"https://doi.org/10.1109/CSCloud-EdgeCom58631.2023.00037","url":null,"abstract":"Conducting binary function naming helps reverse engineers understand the internal workings of the code and perform malicious code analysis without accessing the source code. However, the loss of debugging information poses the challenge of insufficient high-level semantic information description for stripping binary code function naming. Meanwhile, the existing binary function naming scheme has one function label for only one sample. The long-tail effect of function labels for a single sample makes the machine learning-based prediction models face the challenge. To obtain a function correlation label and improve the propensity score of uncommon tail labels, we propose a multi-label learning-based binary function naming model BContext2Name. This model automatically generates relevant labels for binary function naming by function context information with the help of PfastreXML model. The experimental results show that BContext2Name can enrich function labels and alleviate the long-tail effect that exists for a single sample class. To obtain high-level semantics of binary functions, we align pseudocode and basic blocks based on disassembly and decompilation, identify concrete or abstract values of API parameters by variable tracking, and construct API-enhanced control flow graphs. Finally, a seq2seq neural network translation model with attention mechanism is constructed between function multi-label learning and enhanced control flow graphs. Experiments on the dataset reveal that the F1 values of the BContext2Name model improve by 3.55% and 15.23% over the state-of-the-art XFL and Nero, respectively. This indicates that function multi-label learning can provide accurate labels for binary functions and can help reverse analysts understand the inner working mechanism of binary code. Code and data for this evaluation are available at https://github.com/CSecurityZhongYuan/BContext2Name.","PeriodicalId":56007,"journal":{"name":"Journal of Cloud Computing-Advances Systems and Applications","volume":"22 1","pages":"167-172"},"PeriodicalIF":4.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85100602","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}
In the context of the national big data strategy, physical fitness test data has become one of the main influencing factors in guiding and promoting the participation of the population in sports and fitness. Recommending exercise prescriptions based on national physical fitness test data has become an important research topic. However, currently, there is little research on how to accurately use computer data processing technology to recommend exercise prescriptions based on physical fitness test data. In this study, we propose a ResNet-based Exercise Prescription (ResNet-EP) method that utilizes one-dimensional residual neural network technology to recommend exercise prescriptions based on physical fitness testing data. This method comprehensively analyzes physical fitness testing data and exercise prescription data and realizes the automatic recommendation of exercise prescriptions. Experimental results on a real dataset demonstrate that the ResNet-EP model outperforms other comparison models in terms of precision (79.98%), recall (83.73%), and F1 score (81.81%). This study provides novel insights into the combination of physical fitness testing and exercise.
{"title":"A One-Dimensional Residual Network and Physical Fitness-Based Exercise Prescription Recommendation Method","authors":"Runqing Fan, Zhenlian Peng, Buqing Cao, Jianxun Liu, Peng Che, Tieping Chen","doi":"10.1109/CSCloud-EdgeCom58631.2023.00046","DOIUrl":"https://doi.org/10.1109/CSCloud-EdgeCom58631.2023.00046","url":null,"abstract":"In the context of the national big data strategy, physical fitness test data has become one of the main influencing factors in guiding and promoting the participation of the population in sports and fitness. Recommending exercise prescriptions based on national physical fitness test data has become an important research topic. However, currently, there is little research on how to accurately use computer data processing technology to recommend exercise prescriptions based on physical fitness test data. In this study, we propose a ResNet-based Exercise Prescription (ResNet-EP) method that utilizes one-dimensional residual neural network technology to recommend exercise prescriptions based on physical fitness testing data. This method comprehensively analyzes physical fitness testing data and exercise prescription data and realizes the automatic recommendation of exercise prescriptions. Experimental results on a real dataset demonstrate that the ResNet-EP model outperforms other comparison models in terms of precision (79.98%), recall (83.73%), and F1 score (81.81%). This study provides novel insights into the combination of physical fitness testing and exercise.","PeriodicalId":56007,"journal":{"name":"Journal of Cloud Computing-Advances Systems and Applications","volume":"28 1","pages":"223-228"},"PeriodicalIF":4.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81607808","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}
Pub Date : 2023-07-01DOI: 10.1109/CSCloud-EdgeCom58631.2023.00014
Zisang Xu, Ruirui Zhang, Peng Huang, Jianbo Xu
In the Internet of Vehicles(IoV) based on federated learning, the vehicle avoids the server from collecting sensitive data of users by uploading model parameters. However, after research, it is found that the model parameters uploaded by the vehicle are also vulnerable to model inversion attacks or other attacks, thus exposing sensitive data of users. Therefore, this paper proposes an aggregation protocol resisting collusion attacks in the Internet of Vehicles environment. First, the Roadside Unit (RSU) and the Trusted Authority (TA) cooperate to issue tokens for the vehicle to reduce the authentication overhead of the vehicle frequently crossing domains. Second, the protocol uses blinding factors and secret sharing techniques to effectively resist collusion attacks between entities. Finally, after mathematical analysis, it is proved that the protocol has high security and efficiency.
{"title":"An Aggregation Protocol Resisting Collusion Attacks in the Internet of Vehicles Environment","authors":"Zisang Xu, Ruirui Zhang, Peng Huang, Jianbo Xu","doi":"10.1109/CSCloud-EdgeCom58631.2023.00014","DOIUrl":"https://doi.org/10.1109/CSCloud-EdgeCom58631.2023.00014","url":null,"abstract":"In the Internet of Vehicles(IoV) based on federated learning, the vehicle avoids the server from collecting sensitive data of users by uploading model parameters. However, after research, it is found that the model parameters uploaded by the vehicle are also vulnerable to model inversion attacks or other attacks, thus exposing sensitive data of users. Therefore, this paper proposes an aggregation protocol resisting collusion attacks in the Internet of Vehicles environment. First, the Roadside Unit (RSU) and the Trusted Authority (TA) cooperate to issue tokens for the vehicle to reduce the authentication overhead of the vehicle frequently crossing domains. Second, the protocol uses blinding factors and secret sharing techniques to effectively resist collusion attacks between entities. Finally, after mathematical analysis, it is proved that the protocol has high security and efficiency.","PeriodicalId":56007,"journal":{"name":"Journal of Cloud Computing-Advances Systems and Applications","volume":"35 2 1","pages":"24-29"},"PeriodicalIF":4.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77691398","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：