Pub Date : 2023-09-02DOI: 10.1080/1206212X.2023.2252238
J. Odoom, Xiaofang Huang, Zuhong Zhou, S. Danso, Benedicta Nana Esi Nyarko, Jinan Zheng, Yanjie Xiang
The quest to share Electronic Heath Records (EHR) with blockchain as a core technology has witnessed a myriad of frameworks ingrained with diverse cryptographic primitives as well as non-cryptographic mechanisms. Existing works, however, still suffer from privacy-related challenges chief being privacy breaches based on blockchain digital footprints from health facilities, doctors, and patients alike as well as feasibility challenges. Empirical research from state-of-the-art demonstrates the possibility to deanonymize entities involved in a blockchain transaction via inference analysis using such digital footprints on-chain. In this paper, we address such lacunae by advancing a privacy-conscious feasible blockchain-agnostic EHR sharing framework leveraging anonymous transactions, a smart contract, and decentralized storage technology. We construct a constant-size identity-based ring signature to provide accentuated privacy for transaction initiators and demonstrate how health facilities can anonymously retrieve anonymous data on-chain to facilitate EHR sharing via a novel, robust yet computationally efficient, and privacy-aware algorithm dubbed PatientFinder. We subsequently show proof of concept of our framework. A thorough system evaluation is performed revealing that the solution satisfies the privacy of patients and health facilities (doctors), feasibility, and security-related requirements.
{"title":"Blockchain-assisted sharing of electronic health records: a feasible privacy-centric constant-size ring signature framework","authors":"J. Odoom, Xiaofang Huang, Zuhong Zhou, S. Danso, Benedicta Nana Esi Nyarko, Jinan Zheng, Yanjie Xiang","doi":"10.1080/1206212X.2023.2252238","DOIUrl":"https://doi.org/10.1080/1206212X.2023.2252238","url":null,"abstract":"The quest to share Electronic Heath Records (EHR) with blockchain as a core technology has witnessed a myriad of frameworks ingrained with diverse cryptographic primitives as well as non-cryptographic mechanisms. Existing works, however, still suffer from privacy-related challenges chief being privacy breaches based on blockchain digital footprints from health facilities, doctors, and patients alike as well as feasibility challenges. Empirical research from state-of-the-art demonstrates the possibility to deanonymize entities involved in a blockchain transaction via inference analysis using such digital footprints on-chain. In this paper, we address such lacunae by advancing a privacy-conscious feasible blockchain-agnostic EHR sharing framework leveraging anonymous transactions, a smart contract, and decentralized storage technology. We construct a constant-size identity-based ring signature to provide accentuated privacy for transaction initiators and demonstrate how health facilities can anonymously retrieve anonymous data on-chain to facilitate EHR sharing via a novel, robust yet computationally efficient, and privacy-aware algorithm dubbed PatientFinder. We subsequently show proof of concept of our framework. A thorough system evaluation is performed revealing that the solution satisfies the privacy of patients and health facilities (doctors), feasibility, and security-related requirements.","PeriodicalId":39673,"journal":{"name":"International Journal of Computers and Applications","volume":"21 1","pages":"564 - 578"},"PeriodicalIF":0.0,"publicationDate":"2023-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82504363","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-02DOI: 10.1080/1206212X.2023.2250143
Samar Bouazizi, Emna Benmohamed, Hela Ltifi
Bayesian networks, revered for their adeptness in modeling uncertainty and predicting outcomes, encounter a formidable hurdle during the structure learning phase – an NP-hard problem, posing insurmountable computational challenges for large networks. To surmount this barrier and advance the field, we propose an innovative optimization of the K2PC algorithm for Bayesian network structure learning. Derived from the popular K2 algorithm, our novel optimization ingeniously tackles K2PC's vulnerability to predetermined node order. Leveraging the power of a particle swarm optimization algorithm, we adeptly seek the optimal node ordering, yielding exceptional results. Through rigorous evaluations on benchmark networks, our proposed method surpasses prior approaches in structure difference and accuracy, affirming its potential as a promising avenue for Bayesian network structure learning in large, complex networks. We posit that our novel approach constitutes an important advance in the field of Bayesian network structure learning, with the potential to stimulate additional progress through further scientific investigation.
{"title":"PSO-K2PC: Bayesian structure learning using optimized K2 algorithm for parents-children detection","authors":"Samar Bouazizi, Emna Benmohamed, Hela Ltifi","doi":"10.1080/1206212X.2023.2250143","DOIUrl":"https://doi.org/10.1080/1206212X.2023.2250143","url":null,"abstract":"Bayesian networks, revered for their adeptness in modeling uncertainty and predicting outcomes, encounter a formidable hurdle during the structure learning phase – an NP-hard problem, posing insurmountable computational challenges for large networks. To surmount this barrier and advance the field, we propose an innovative optimization of the K2PC algorithm for Bayesian network structure learning. Derived from the popular K2 algorithm, our novel optimization ingeniously tackles K2PC's vulnerability to predetermined node order. Leveraging the power of a particle swarm optimization algorithm, we adeptly seek the optimal node ordering, yielding exceptional results. Through rigorous evaluations on benchmark networks, our proposed method surpasses prior approaches in structure difference and accuracy, affirming its potential as a promising avenue for Bayesian network structure learning in large, complex networks. We posit that our novel approach constitutes an important advance in the field of Bayesian network structure learning, with the potential to stimulate additional progress through further scientific investigation.","PeriodicalId":39673,"journal":{"name":"International Journal of Computers and Applications","volume":"1 1","pages":"553 - 563"},"PeriodicalIF":0.0,"publicationDate":"2023-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83425579","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-08-03DOI: 10.1080/1206212X.2023.2243676
Shinzeer C. K., Avinash Bhagat, A. Kushwaha
Governments and individuals have taken extraordinary measures to protect the health of the people during the COVID pandemic. Stored medical data remains the main target for hackers, and hence it needs to be stored securely. To achieve this objective, this paper proposes a novel model using Delegated Proof of Stake-Hyper ledger Fabric Block Chain (DPOS-HFBC). Primarily, by employing LL Subbandeigen Value decomposition employed Discrete Wavelet Transform (L2-DWT), the patient’s Lung Computed Tomography (CT) image data are collected and embedded. For embedding, the patient’s name and ID are taken. In embedding, a Pseudorandom number generator using the Mersenne twister algorithm employed in Elliptic Curve Cryptography (PM-ECC) is applied for key encryption. It covered the image that was embedded with the original and then stored in DPOS-HFBC. Likewise, for authorization, every patient’s biometric ID was hashed and stored in DPOS-HFBC. Data requesters request data in the Interplanetary File System (IPFS) of DPOS-HFBC, and the attributes from the request are extracted and sent to the authority for verification. After verifying, the authority shares their biometric ID with the requester and this gets hashed and then verified in DPOS-HFBC. To show the model’s supremacy, the proposed method was evaluated and compared with existing methods.
{"title":"Secure medical data storage in DPOS-hyper ledger fabric block chain using PM-ECC and L2-DWT","authors":"Shinzeer C. K., Avinash Bhagat, A. Kushwaha","doi":"10.1080/1206212X.2023.2243676","DOIUrl":"https://doi.org/10.1080/1206212X.2023.2243676","url":null,"abstract":"Governments and individuals have taken extraordinary measures to protect the health of the people during the COVID pandemic. Stored medical data remains the main target for hackers, and hence it needs to be stored securely. To achieve this objective, this paper proposes a novel model using Delegated Proof of Stake-Hyper ledger Fabric Block Chain (DPOS-HFBC). Primarily, by employing LL Subbandeigen Value decomposition employed Discrete Wavelet Transform (L2-DWT), the patient’s Lung Computed Tomography (CT) image data are collected and embedded. For embedding, the patient’s name and ID are taken. In embedding, a Pseudorandom number generator using the Mersenne twister algorithm employed in Elliptic Curve Cryptography (PM-ECC) is applied for key encryption. It covered the image that was embedded with the original and then stored in DPOS-HFBC. Likewise, for authorization, every patient’s biometric ID was hashed and stored in DPOS-HFBC. Data requesters request data in the Interplanetary File System (IPFS) of DPOS-HFBC, and the attributes from the request are extracted and sent to the authority for verification. After verifying, the authority shares their biometric ID with the requester and this gets hashed and then verified in DPOS-HFBC. To show the model’s supremacy, the proposed method was evaluated and compared with existing methods.","PeriodicalId":39673,"journal":{"name":"International Journal of Computers and Applications","volume":"103 1","pages":"516 - 522"},"PeriodicalIF":0.0,"publicationDate":"2023-08-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79456098","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-08-03DOI: 10.1080/1206212X.2023.2252117
Farid Feyzi, Arman Daneshdoost
Accurate prediction of defective software modules is of great importance for prioritizing quality assurance efforts, reasonably allocating testing resources, reducing costs and improving software quality. Several studies have used machine learning to predict software defects. However, complex structures and imbalanced class distributions in software defect data make learning an effective defect prediction model challenging. In this article, two deep learning-based defect prediction models using static code metrics are proposed. In order to enhance the learning process and improve the performance of the proposed models, pool-based active learning is employed. In this regard, the possibility of using active learning to mitigate the need for a large amount of labeled data in the process of building deep learning models is investigated. To deal with imbalanced distribution of software modules between defective and non-defective classes, Near-Miss under-sampling and KNN, with different number of neighbors, are used. The reason for choosing them is their good performance in binary classification problems. Experiments are performed on two well-known, publicly available datasets, GitHub Bug Dataset and public Unified Bug Dataset for java projects. The evaluation results reveal the effectiveness of our proposed models in comparison to the traditional machine learning algorithms. In the conducted investigations on the Unified Bug Dataset, at the file level, the value of F-measure and AUC criteria have improved by 13 and 11 percent, respectively and at the class level, the values have improved by 14 and 11 percent, respectively.
{"title":"Studying the effectiveness of deep active learning in software defect prediction","authors":"Farid Feyzi, Arman Daneshdoost","doi":"10.1080/1206212X.2023.2252117","DOIUrl":"https://doi.org/10.1080/1206212X.2023.2252117","url":null,"abstract":"Accurate prediction of defective software modules is of great importance for prioritizing quality assurance efforts, reasonably allocating testing resources, reducing costs and improving software quality. Several studies have used machine learning to predict software defects. However, complex structures and imbalanced class distributions in software defect data make learning an effective defect prediction model challenging. In this article, two deep learning-based defect prediction models using static code metrics are proposed. In order to enhance the learning process and improve the performance of the proposed models, pool-based active learning is employed. In this regard, the possibility of using active learning to mitigate the need for a large amount of labeled data in the process of building deep learning models is investigated. To deal with imbalanced distribution of software modules between defective and non-defective classes, Near-Miss under-sampling and KNN, with different number of neighbors, are used. The reason for choosing them is their good performance in binary classification problems. Experiments are performed on two well-known, publicly available datasets, GitHub Bug Dataset and public Unified Bug Dataset for java projects. The evaluation results reveal the effectiveness of our proposed models in comparison to the traditional machine learning algorithms. In the conducted investigations on the Unified Bug Dataset, at the file level, the value of F-measure and AUC criteria have improved by 13 and 11 percent, respectively and at the class level, the values have improved by 14 and 11 percent, respectively.","PeriodicalId":39673,"journal":{"name":"International Journal of Computers and Applications","volume":"16 1","pages":"534 - 552"},"PeriodicalIF":0.0,"publicationDate":"2023-08-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81283313","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-08-03DOI: 10.1080/1206212X.2023.2250149
Shobha Chawla, N. Gupta
Outsourcing massive amounts of data to the cloud service provider (CSP) has raised various security concerns for data confidentiality and access control. The ciphertext policy attribute based encryption (CPABE) scheme allows data owners to impose access control on their cloud-resident sensitive data. This paper has studied the approaches adopted to revoke users by the existing bilinear pairing cryptography based CPABE schemes. The existing studies have suggested solutions to revocation either by updating the non-revoked users’ keys or updating the ciphertext. Such approaches increase computational overhead for resource-constrained devices. In addition, a few studies have discussed the possibility of the CSP becoming dishonest and colluding with the revoked users. The likelihood of a collusion attack caused by the CSP and the revoked users also needs extensive attention. The development of the proposed proxy-based framework aims to extend the existing CPABE scheme and simplify the revocation of access rights at the user and attribute level with scalability, dynamicity, collusion resistance, and forward/backward secrecy. The proposed framework uses bilinear pairing cryptography and LSSS as an access structure. Furthermore, the security and performance analysis of the proposed framework reflects that it is implementable, better, and more secure than the existing work.
{"title":"A cloud based enhanced CPABE framework for efficient user and attribute-level revocation","authors":"Shobha Chawla, N. Gupta","doi":"10.1080/1206212X.2023.2250149","DOIUrl":"https://doi.org/10.1080/1206212X.2023.2250149","url":null,"abstract":"Outsourcing massive amounts of data to the cloud service provider (CSP) has raised various security concerns for data confidentiality and access control. The ciphertext policy attribute based encryption (CPABE) scheme allows data owners to impose access control on their cloud-resident sensitive data. This paper has studied the approaches adopted to revoke users by the existing bilinear pairing cryptography based CPABE schemes. The existing studies have suggested solutions to revocation either by updating the non-revoked users’ keys or updating the ciphertext. Such approaches increase computational overhead for resource-constrained devices. In addition, a few studies have discussed the possibility of the CSP becoming dishonest and colluding with the revoked users. The likelihood of a collusion attack caused by the CSP and the revoked users also needs extensive attention. The development of the proposed proxy-based framework aims to extend the existing CPABE scheme and simplify the revocation of access rights at the user and attribute level with scalability, dynamicity, collusion resistance, and forward/backward secrecy. The proposed framework uses bilinear pairing cryptography and LSSS as an access structure. Furthermore, the security and performance analysis of the proposed framework reflects that it is implementable, better, and more secure than the existing work.","PeriodicalId":39673,"journal":{"name":"International Journal of Computers and Applications","volume":"42 1","pages":"523 - 533"},"PeriodicalIF":0.0,"publicationDate":"2023-08-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"91333103","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper presents a traffic rerouting approach to keep a satisfactory QoS (Quality of Service) in virtualized network infrastructures (VPN, Cloud, etc.), supervised by an SDN (Software-Defined Networking) controller dealing with high traffic fluctuations. Traffic fluctuations are usually caused by link failures or link congestion and result in high packet loss, long delay times, and high jitter. These metrics are critical in computer network's QoS assessing. In our strategy, we combine traffic prediction in the SDN controller with flow splitting in the data plane. Simulations reveal that this strategy provides more satisfying values of the QoS assessment metrics compared to the literature.
{"title":"Traffic predictive-based flow splitting rerouting scheme for link failures in software-defined networks","authors":"Vianney Kengne Tchendji, Joelle Kabdjou, Yannick Florian Yankam","doi":"10.1080/1206212X.2023.2241185","DOIUrl":"https://doi.org/10.1080/1206212X.2023.2241185","url":null,"abstract":"This paper presents a traffic rerouting approach to keep a satisfactory QoS (Quality of Service) in virtualized network infrastructures (VPN, Cloud, etc.), supervised by an SDN (Software-Defined Networking) controller dealing with high traffic fluctuations. Traffic fluctuations are usually caused by link failures or link congestion and result in high packet loss, long delay times, and high jitter. These metrics are critical in computer network's QoS assessing. In our strategy, we combine traffic prediction in the SDN controller with flow splitting in the data plane. Simulations reveal that this strategy provides more satisfying values of the QoS assessment metrics compared to the literature.","PeriodicalId":39673,"journal":{"name":"International Journal of Computers and Applications","volume":"81 1","pages":"508 - 515"},"PeriodicalIF":0.0,"publicationDate":"2023-07-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90037903","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-07-28DOI: 10.1080/1206212X.2023.2239551
Chigurupati Ravi Swaroop, K. Raja
In recent years, outlier detection has attained great attention with machine learning techniques due to its wide range of applications. By considering the input data’s distributive nature and large dimensionality, outlier detection becomes a challenging issue. Robust outlier detection systems are crucial for data pattern prediction without labeled data. This research develops a novel approach based on stacking auto encoders over Long-Short Term Memory (LSTM) for outlier prediction. The detection accuracy of outlier detection is improved with the hyperparameters optimized with the Chaotic Gravitational Search Algorithm (CGSA). CGSA minimizes the training loss with enhanced detection accuracy in the proposed outlier detection process. The auto encoder in outlier detection transforms the input into a latent space representation to generate the original input sequence. The involvement of learning parameters computes and minimizes the errors between input and generated sequences. The proposed work is experimented and compared with state-of-the-art approaches of recent research. Using the proposed approach, the performance of outlier prediction is improved with an accuracy of98.6%, sensitivity of 96.1%, specificity of 97.8%, G-mean of 96%, Area Under Curve (AUC) of 0.935, Hit rate of 92.3%. Also, the outlier detection errors are minimized, showing the proposed approach’s efficiency.
{"title":"CGSA optimized LSTM auto encoder for outlier detection","authors":"Chigurupati Ravi Swaroop, K. Raja","doi":"10.1080/1206212X.2023.2239551","DOIUrl":"https://doi.org/10.1080/1206212X.2023.2239551","url":null,"abstract":"In recent years, outlier detection has attained great attention with machine learning techniques due to its wide range of applications. By considering the input data’s distributive nature and large dimensionality, outlier detection becomes a challenging issue. Robust outlier detection systems are crucial for data pattern prediction without labeled data. This research develops a novel approach based on stacking auto encoders over Long-Short Term Memory (LSTM) for outlier prediction. The detection accuracy of outlier detection is improved with the hyperparameters optimized with the Chaotic Gravitational Search Algorithm (CGSA). CGSA minimizes the training loss with enhanced detection accuracy in the proposed outlier detection process. The auto encoder in outlier detection transforms the input into a latent space representation to generate the original input sequence. The involvement of learning parameters computes and minimizes the errors between input and generated sequences. The proposed work is experimented and compared with state-of-the-art approaches of recent research. Using the proposed approach, the performance of outlier prediction is improved with an accuracy of98.6%, sensitivity of 96.1%, specificity of 97.8%, G-mean of 96%, Area Under Curve (AUC) of 0.935, Hit rate of 92.3%. Also, the outlier detection errors are minimized, showing the proposed approach’s efficiency.","PeriodicalId":39673,"journal":{"name":"International Journal of Computers and Applications","volume":"223 1","pages":"497 - 507"},"PeriodicalIF":0.0,"publicationDate":"2023-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75553353","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-07-25DOI: 10.1080/1206212X.2023.2237766
Fjolla Berisha, E. Bytyçi
Filtering information on the Internet and recommending the right choices is more than important for Internet users and various businesses that offer products and services. Although recommender systems do this work efficiently, problems such as Cold Start often appear when new users or items enter the system. The traditional methods of recommender systems, collaborative filtering and content–based techniques, do not offer an optimized solution to this problem. The integration of neural networks in recommender systems offers a new approach to solving cold start. Whether using the feature of extracting hidden data, or using deep learning algorithms with more layers, the accuracy of recommendations and predictions has increased significantly. We have analyzed 40 papers that approached solving the cold start problem using neural networks. We have researched how neural networks are integrated into recommender systems, what they are used for, which neural network algorithms have shown to be more efficient in solving the cold start problem, and which algorithms have increased the accuracy of the recommendation. We aim to answer these questions with other subquestions related to types of cold start such as item or user cold start and warm, partial, or strict cold start.
{"title":"Addressing cold start in recommender systems with neural networks: a literature survey","authors":"Fjolla Berisha, E. Bytyçi","doi":"10.1080/1206212X.2023.2237766","DOIUrl":"https://doi.org/10.1080/1206212X.2023.2237766","url":null,"abstract":"Filtering information on the Internet and recommending the right choices is more than important for Internet users and various businesses that offer products and services. Although recommender systems do this work efficiently, problems such as Cold Start often appear when new users or items enter the system. The traditional methods of recommender systems, collaborative filtering and content–based techniques, do not offer an optimized solution to this problem. The integration of neural networks in recommender systems offers a new approach to solving cold start. Whether using the feature of extracting hidden data, or using deep learning algorithms with more layers, the accuracy of recommendations and predictions has increased significantly. We have analyzed 40 papers that approached solving the cold start problem using neural networks. We have researched how neural networks are integrated into recommender systems, what they are used for, which neural network algorithms have shown to be more efficient in solving the cold start problem, and which algorithms have increased the accuracy of the recommendation. We aim to answer these questions with other subquestions related to types of cold start such as item or user cold start and warm, partial, or strict cold start.","PeriodicalId":39673,"journal":{"name":"International Journal of Computers and Applications","volume":"95 1","pages":"485 - 496"},"PeriodicalIF":0.0,"publicationDate":"2023-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81853183","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-03DOI: 10.1080/1206212X.2023.2232169
Fang Duan
The existing behavior recognition models based on the deep convolutional neural network have some problems, such as feature extraction with a single scale and insufficient feature utilization in the middle level. In this paper, we propose a multi-scale residual aggregation feature network based on multi-time division for behavior recognition. Through the sampling form of multi-time division, the diversity of behavior depth features is enriched. Firstly, a hybrid extended convolution residual block (HERB) is designed using extended convolution and residual join with different extension coefficients to extract feature information at multiple scales effectively. Secondly, a feature aggregation mechanism (AM) is introduced to solve the problem of insufficient feature utilization in the middle layer of the network. We construct a deep aggregation model that can learn the distribution of complex behavior features to solve the problem of human behavior classification over a long time span. Experiments on behavioral datasets UCF101 and HMDB51 verify the effectiveness of the new algorithm.
{"title":"Multi-scale residual aggregation feature network based on multi-time division for motion behavior recognition","authors":"Fang Duan","doi":"10.1080/1206212X.2023.2232169","DOIUrl":"https://doi.org/10.1080/1206212X.2023.2232169","url":null,"abstract":"The existing behavior recognition models based on the deep convolutional neural network have some problems, such as feature extraction with a single scale and insufficient feature utilization in the middle level. In this paper, we propose a multi-scale residual aggregation feature network based on multi-time division for behavior recognition. Through the sampling form of multi-time division, the diversity of behavior depth features is enriched. Firstly, a hybrid extended convolution residual block (HERB) is designed using extended convolution and residual join with different extension coefficients to extract feature information at multiple scales effectively. Secondly, a feature aggregation mechanism (AM) is introduced to solve the problem of insufficient feature utilization in the middle layer of the network. We construct a deep aggregation model that can learn the distribution of complex behavior features to solve the problem of human behavior classification over a long time span. Experiments on behavioral datasets UCF101 and HMDB51 verify the effectiveness of the new algorithm.","PeriodicalId":39673,"journal":{"name":"International Journal of Computers and Applications","volume":"20 1","pages":"452 - 459"},"PeriodicalIF":0.0,"publicationDate":"2023-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84673500","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-03DOI: 10.1080/1206212X.2023.2232167
A. Touzene, Manar Al Moqbali
This paper presents a new load shifting strategy for smart grid systems based on both power consumers’ day-ahead power forecast and their Service Level Agreement (SLA) in order to reduce their electricity bills, guaranties user satisfaction, and for smart grid system to reduce as well the overall power consumption at the peak hours. We provide an analytical model that formulated the load shifting process as a cost minimization problem. A Genetic Algorithm (GA) approach based on a two dimensional chromosome representation is used to solve the optimization problem by collecting a day-ahead forecast and SLAs as an input from the power consumers. The output of the GA consists of giving the best power task plan for the day-ahead which satisfy all consumers in terms of minimizing their consumption bill and reduces the peak demand. Experimental results using simulation show that the proposed load shifting strategy not only guaranty SLA requirements but it reduces the total cost by more than 16%, and in general it achieves a substantial cost savings of 38% compared to the recent algorithms from the literature.
{"title":"User satisfaction-based genetic algorithm for load shifting in smart grid","authors":"A. Touzene, Manar Al Moqbali","doi":"10.1080/1206212X.2023.2232167","DOIUrl":"https://doi.org/10.1080/1206212X.2023.2232167","url":null,"abstract":"This paper presents a new load shifting strategy for smart grid systems based on both power consumers’ day-ahead power forecast and their Service Level Agreement (SLA) in order to reduce their electricity bills, guaranties user satisfaction, and for smart grid system to reduce as well the overall power consumption at the peak hours. We provide an analytical model that formulated the load shifting process as a cost minimization problem. A Genetic Algorithm (GA) approach based on a two dimensional chromosome representation is used to solve the optimization problem by collecting a day-ahead forecast and SLAs as an input from the power consumers. The output of the GA consists of giving the best power task plan for the day-ahead which satisfy all consumers in terms of minimizing their consumption bill and reduces the peak demand. Experimental results using simulation show that the proposed load shifting strategy not only guaranty SLA requirements but it reduces the total cost by more than 16%, and in general it achieves a substantial cost savings of 38% compared to the recent algorithms from the literature.","PeriodicalId":39673,"journal":{"name":"International Journal of Computers and Applications","volume":"1 1","pages":"444 - 451"},"PeriodicalIF":0.0,"publicationDate":"2023-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84159562","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
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：