Pub Date : 2023-06-10DOI: 10.37936/ecti-cit.2023172.251440
Vara Varavithya, Supakit Prueksaaroon
In the last decade, government organizations and private companies in Thailand have invested considerably in computing resources. Research collaborations typically band together based on shared interests and propose projects to compete for funding. Thus, many institutions frequently use similar computing resources. A bird's eye view of high-performance computer infrastructure in Thailand is important in many ways. To the best of our knowledge and ability, we gathered information on government procurements of HPC resources in the past five years, which cover several organizations and ministries. We list the system specifications and tabulate the target application areas for each system. The aggregated number of cores and storage space for HPC in Thailand, commissioned during the past five years, is 54,838 cores and 21 PB, respectively. We also describe the large data transfers using UniNet for HPC applications. The survey results can be used by academics and decision-makers to build research agendas and national development strategies.
{"title":"A Survey of High Performance Computing (HPC) Infrastructure in Thailand","authors":"Vara Varavithya, Supakit Prueksaaroon","doi":"10.37936/ecti-cit.2023172.251440","DOIUrl":"https://doi.org/10.37936/ecti-cit.2023172.251440","url":null,"abstract":"In the last decade, government organizations and private companies in Thailand have invested considerably in computing resources. Research collaborations typically band together based on shared interests and propose projects to compete for funding. Thus, many institutions frequently use similar computing resources. A bird's eye view of high-performance computer infrastructure in Thailand is important in many ways. To the best of our knowledge and ability, we gathered information on government procurements of HPC resources in the past five years, which cover several organizations and ministries. We list the system specifications and tabulate the target application areas for each system. The aggregated number of cores and storage space for HPC in Thailand, commissioned during the past five years, is 54,838 cores and 21 PB, respectively. We also describe the large data transfers using UniNet for HPC applications. The survey results can be used by academics and decision-makers to build research agendas and national development strategies.","PeriodicalId":38808,"journal":{"name":"Transactions on Electrical Engineering, Electronics, and Communications","volume":"125 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-06-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85276223","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 : 2023-06-08DOI: 10.37936/ecti-cit.2023172.252270
S. Fugkeaw, Pattavee Sanchol
Personal data leakage prevention has now become a critical issue for implementing data management and sharing in many industries. Several data privacy regulations such as General Data Protection Regulation (GDPR), Health Insurance Portability and Accountability Act (HIPPA), California Consumer Privacy Act (CCPA), and Thailand's Personal Data Protection Act (PDPA) have been issued to enforce organizations to collect, process, and transfer personally identifiable information (PII) securely. In this paper, we propose a design and development of PII RapidDiscover, an efficient Thai and English PII discovery system featured with automatic consent discovery. At the core of our proposed system, we introduce the PII scanning algorithm based on the Presidio library and a natural language processing (NLP) technique to improve the scan result of PII written in Thai and English. Finally, we conducted the experiments to demonstrate the efficiency of our proposed system.
{"title":"Enabling Efficient Personally Identifiable Information Detection with Automatic Consent Discovery","authors":"S. Fugkeaw, Pattavee Sanchol","doi":"10.37936/ecti-cit.2023172.252270","DOIUrl":"https://doi.org/10.37936/ecti-cit.2023172.252270","url":null,"abstract":"Personal data leakage prevention has now become a critical issue for implementing data management and sharing in many industries. Several data privacy regulations such as General Data Protection Regulation (GDPR), Health Insurance Portability and Accountability Act (HIPPA), California Consumer Privacy Act (CCPA), and Thailand's Personal Data Protection Act (PDPA) have been issued to enforce organizations to collect, process, and transfer personally identifiable information (PII) securely. In this paper, we propose a design and development of PII RapidDiscover, an efficient Thai and English PII discovery system featured with automatic consent discovery. At the core of our proposed system, we introduce the PII scanning algorithm based on the Presidio library and a natural language processing (NLP) technique to improve the scan result of PII written in Thai and English. Finally, we conducted the experiments to demonstrate the efficiency of our proposed system.","PeriodicalId":38808,"journal":{"name":"Transactions on Electrical Engineering, Electronics, and Communications","volume":"40 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-06-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78105395","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: 10.37936/ecti-cit.2023172.252014
P. Rani, S. Priya
The most crucial properties of decentralized, immutable blockchain technologies are being transparent, tamper-proof, and have total traceability. With an increase in overseas students globally, the problem of diploma forgery, and the sale of forged credentials, the management and dissemination of student educational information continue to encounter several problems. Privacy violation issues like security, privacy, trustworthiness, consistency challenges, and traceability issues are considered when managing student academic records in educational sectors. The proposed work is a novel decentralized Chameleon Hash Function and it is applied to overcome these privacy violation issues. Security-aware and privacy-preserving blockchain Chameleon hash functions for Education System are suggested because every redaction needs to be approved by numerous blockchain nodes. A Proof of Continuous Work (PoCW) consensus algorithm entirely based on Blockchain is proposed for data storage and sharing, which minimizes processing power wastage to enhance the accessibility and transparency of the procedure for students receiving educational degree certificates. By consistently giving proof of storage, miners can gain an edge in the mining procedure. Without any outside help, Blockchain has created a reliable blockchain-based storage system that does not depend on a third party. The simulation and theoretical study's findings demonstrate that the proposed scheme has enhanced security, trustworthiness, and traceability.
{"title":"Security-Aware and Privacy-Preserving Blockchain Chameleon Hash Functions for Education System","authors":"P. Rani, S. Priya","doi":"10.37936/ecti-cit.2023172.252014","DOIUrl":"https://doi.org/10.37936/ecti-cit.2023172.252014","url":null,"abstract":"The most crucial properties of decentralized, immutable blockchain technologies are being transparent, tamper-proof, and have total traceability. With an increase in overseas students globally, the problem of diploma forgery, and the sale of forged credentials, the management and dissemination of student educational information continue to encounter several problems. Privacy violation issues like security, privacy, trustworthiness, consistency challenges, and traceability issues are considered when managing student academic records in educational sectors. The proposed work is a novel decentralized Chameleon Hash Function and it is applied to overcome these privacy violation issues. Security-aware and privacy-preserving blockchain Chameleon hash functions for Education System are suggested because every redaction needs to be approved by numerous blockchain nodes. A Proof of Continuous Work (PoCW) consensus algorithm entirely based on Blockchain is proposed for data storage and sharing, which minimizes processing power wastage to enhance the accessibility and transparency of the procedure for students receiving educational degree certificates. By consistently giving proof of storage, miners can gain an edge in the mining procedure. Without any outside help, Blockchain has created a reliable blockchain-based storage system that does not depend on a third party. The simulation and theoretical study's findings demonstrate that the proposed scheme has enhanced security, trustworthiness, and traceability.","PeriodicalId":38808,"journal":{"name":"Transactions on Electrical Engineering, Electronics, and Communications","volume":"21 5 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80353812","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 : 2023-05-13DOI: 10.37936/ecti-cit.2023172.251536
Wu Xiaogang, T. Tanprasert
Attention mechanisms in deep learning can focus on critical features and ignore irrelevant details in the target task. This paper proposes a new multi-grained attention model (MGAN) to extract parts from images. The model includes a multi-grain spatial attention (MSA) mechanism and a multi-grain channel attention (MCA) mechanism. We use different convolutional branches and pooling layers to focus on the crucial information in the sample feature space and extract richer multi-grain features from the image. The model uses ResNet and Res2Net as the backbone networks to implement the image classification task. Experiments on the CIFAR10/100 and Mini-Imagenet datasets show that the proposed model MGAN can better focus on the critical information in the sample feature space, extract richer multi-grain features from the images, and significantly improve the image classification accuracy of the network.
{"title":"A Multi-Grained Attention Residual Network for Image Classification","authors":"Wu Xiaogang, T. Tanprasert","doi":"10.37936/ecti-cit.2023172.251536","DOIUrl":"https://doi.org/10.37936/ecti-cit.2023172.251536","url":null,"abstract":"Attention mechanisms in deep learning can focus on critical features and ignore irrelevant details in the target task. This paper proposes a new multi-grained attention model (MGAN) to extract parts from images. The model includes a multi-grain spatial attention (MSA) mechanism and a multi-grain channel attention (MCA) mechanism. We use different convolutional branches and pooling layers to focus on the crucial information in the sample feature space and extract richer multi-grain features from the image. The model uses ResNet and Res2Net as the backbone networks to implement the image classification task. Experiments on the CIFAR10/100 and Mini-Imagenet datasets show that the proposed model MGAN can better focus on the critical information in the sample feature space, extract richer multi-grain features from the images, and significantly improve the image classification accuracy of the network.","PeriodicalId":38808,"journal":{"name":"Transactions on Electrical Engineering, Electronics, and Communications","volume":"14 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83384824","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 : 2023-04-11DOI: 10.37936/ecti-cit.2023172.249022
Pitchakron Thippun, Apidet Booranawong, D. Buranapanichkit
Wireless Body Area Networks (WBANs) are a collection of vital and electrical signals measured from various body parts to help analyze therapeutic approaches for patients using wireless data transmission. The significant data has to communicate with collision avoidance to obtain high throughput. In this paper, a hybrid MAC layer communication is implemented between CSMA/CA and TDMA. CSMA/CA communication has been introduced to manage the TDMA sequence of transmissions without a central node. The experimental results in this system implement real wireless devices, TelosB, with the IEEE 802.15.4 standard. We studied the convergence speed of transmission sequence allocation, which was measured in the CSMA/CA period. When the number of nodes is small, the convergence time is slower than a large number of nodes. However, the number of nodes does not affect the number of rounds entering the transmission period. This parameter has been evaluated for the network and energy efficiency in WBANs. Packet delivery ratio, packet numbers, and energy consumption are examined for the different priority-based nodes in the TDMA period. The energy consumption can reduce to 40% for no priority when compared with high priority in the case of a priority-based node.
{"title":"A Priority-based Data Transmission for Energy Efficiency MAC Protocol with Wireless Body Area Networks","authors":"Pitchakron Thippun, Apidet Booranawong, D. Buranapanichkit","doi":"10.37936/ecti-cit.2023172.249022","DOIUrl":"https://doi.org/10.37936/ecti-cit.2023172.249022","url":null,"abstract":"Wireless Body Area Networks (WBANs) are a collection of vital and electrical signals measured from various body parts to help analyze therapeutic approaches for patients using wireless data transmission. The significant data has to communicate with collision avoidance to obtain high throughput. In this paper, a hybrid MAC layer communication is implemented between CSMA/CA and TDMA. CSMA/CA communication has been introduced to manage the TDMA sequence of transmissions without a central node. The experimental results in this system implement real wireless devices, TelosB, with the IEEE 802.15.4 standard. We studied the convergence speed of transmission sequence allocation, which was measured in the CSMA/CA period. When the number of nodes is small, the convergence time is slower than a large number of nodes. However, the number of nodes does not affect the number of rounds entering the transmission period. This parameter has been evaluated for the network and energy efficiency in WBANs. Packet delivery ratio, packet numbers, and energy consumption are examined for the different priority-based nodes in the TDMA period. The energy consumption can reduce to 40% for no priority when compared with high priority in the case of a priority-based node.","PeriodicalId":38808,"journal":{"name":"Transactions on Electrical Engineering, Electronics, and Communications","volume":"39 9 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73913345","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 : 2023-04-08DOI: 10.37936/ecti-cit.2023172.249484
Subhrajit Sinha Roy, A. Basu, A. Chattopadhyay
Telemedicine is one of the most eminent terms used in the modern e-healthcare system. Digital medical image reports, along with electronic patient records, play a central role in diagnosis from distance. These reports need to be transmitted over an open communication channel with immense security and reliability, so that appropriate diagnosis can be performed. Moreover, the privacy of the patient is required to be preserved. Digital watermarking is one of the most conventional and suitable practices to serve all of these purposes. New challenges appear in the domain of watermarking with the advancement of digital signal processing; consequently, researchers are endowing more efforts to overcome these challenges. In this paper, a survey on medical image watermarking has been done, and the performance of a few state-of-the-art medical image watermarking techniques is compared. This work makes the researchers and the developers familiar with the recent trends, challenges, and scopes in this domain to facilitate them in finding out adequate research directions.
{"title":"Prospects of Digital Watermarking in Providing Security, Reliability, and Privacy to Medical Images","authors":"Subhrajit Sinha Roy, A. Basu, A. Chattopadhyay","doi":"10.37936/ecti-cit.2023172.249484","DOIUrl":"https://doi.org/10.37936/ecti-cit.2023172.249484","url":null,"abstract":"Telemedicine is one of the most eminent terms used in the modern e-healthcare system. Digital medical image reports, along with electronic patient records, play a central role in diagnosis from distance. These reports need to be transmitted over an open communication channel with immense security and reliability, so that appropriate diagnosis can be performed. Moreover, the privacy of the patient is required to be preserved. Digital watermarking is one of the most conventional and suitable practices to serve all of these purposes. New challenges appear in the domain of watermarking with the advancement of digital signal processing; consequently, researchers are endowing more efforts to overcome these challenges. In this paper, a survey on medical image watermarking has been done, and the performance of a few state-of-the-art medical image watermarking techniques is compared. This work makes the researchers and the developers familiar with the recent trends, challenges, and scopes in this domain to facilitate them in finding out adequate research directions.","PeriodicalId":38808,"journal":{"name":"Transactions on Electrical Engineering, Electronics, and Communications","volume":"151 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-04-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76460303","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 : 2023-04-01DOI: 10.37936/ecti-cit.2020.49452
Ecti Cit
ECTI-CIT
ECTI-CIT
{"title":"ECTI Transactions Manuscript Format (Word template)","authors":"Ecti Cit","doi":"10.37936/ecti-cit.2020.49452","DOIUrl":"https://doi.org/10.37936/ecti-cit.2020.49452","url":null,"abstract":"ECTI-CIT","PeriodicalId":38808,"journal":{"name":"Transactions on Electrical Engineering, Electronics, and Communications","volume":"14 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87309846","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 research is an application of information technology to efficiently manage the educational information of the School of Dentistry. Dental students must study medical history in teaching dentistry, although the systematic shelving of documents requires greater flexibility in retrieving and searching for patient cases to be studied. The researchers developed a workflow system to increase the efficiency of document borrowing in the form of a web application. to reduce the process of searching for borrowing documents, reduce the waiting time for documents to be borrowed, and shorten the time to summarize all borrowing information, including adding a user-generated hashtag feature to make it easier to find documents by category and context. By using the system to replace the original process, the total time in every process can be reduced from 180–350 minutes to only 8–40 minutes, equivalent to 241 minutes of reduced wasted time.
{"title":"The Borrowing Medical Records System : A Case Study of Medical Information System","authors":"Juthamat Boonkleang, Pantakarn Supapong, Vittayasak Rujivorakul","doi":"10.1109/ECTIDAMTNCON57770.2023.10139673","DOIUrl":"https://doi.org/10.1109/ECTIDAMTNCON57770.2023.10139673","url":null,"abstract":"This research is an application of information technology to efficiently manage the educational information of the School of Dentistry. Dental students must study medical history in teaching dentistry, although the systematic shelving of documents requires greater flexibility in retrieving and searching for patient cases to be studied. The researchers developed a workflow system to increase the efficiency of document borrowing in the form of a web application. to reduce the process of searching for borrowing documents, reduce the waiting time for documents to be borrowed, and shorten the time to summarize all borrowing information, including adding a user-generated hashtag feature to make it easier to find documents by category and context. By using the system to replace the original process, the total time in every process can be reduced from 180–350 minutes to only 8–40 minutes, equivalent to 241 minutes of reduced wasted time.","PeriodicalId":38808,"journal":{"name":"Transactions on Electrical Engineering, Electronics, and Communications","volume":"413 1","pages":"69-74"},"PeriodicalIF":0.0,"publicationDate":"2023-03-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73051119","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 : 2023-03-22DOI: 10.1109/ECTIDAMTNCON57770.2023.10139616
Jirawit Yanchinda, Fanke Xu
The disruption of the supply chain happens more frequently than in the past; meanwhile the organization is struggling with insufficient digital transformation for supply chain resilience. This research is going to conduct an ontology from the journal text to enable enterprise transfer to digital transformation. Use the database of the Web of science core collection; based on the R package, and we do the text mining from the top-20 high citation article to filter out related terms from the perspective of tf and tf-idf, manually select the ontologies to build up a roadmap integrated a three-line defense of supply chain resilience to matches the before, during, after disruption stages.
供应链的中断比过去发生得更频繁;与此同时，该组织正在努力解决供应链弹性的数字化转型不足的问题。本研究将对期刊文本进行本体化，使企业向数字化转型。利用Web of science核心馆藏数据库;基于R包，我们从前20位高引用文章中进行文本挖掘，从tf和tf-idf的角度过滤出相关术语，手动选择本体构建一个路线图，集成了供应链弹性的三线防御，以匹配中断之前，期间和之后的阶段。
{"title":"Ontology Creation based on Digital Transformation for Supply Chain Resilience","authors":"Jirawit Yanchinda, Fanke Xu","doi":"10.1109/ECTIDAMTNCON57770.2023.10139616","DOIUrl":"https://doi.org/10.1109/ECTIDAMTNCON57770.2023.10139616","url":null,"abstract":"The disruption of the supply chain happens more frequently than in the past; meanwhile the organization is struggling with insufficient digital transformation for supply chain resilience. This research is going to conduct an ontology from the journal text to enable enterprise transfer to digital transformation. Use the database of the Web of science core collection; based on the R package, and we do the text mining from the top-20 high citation article to filter out related terms from the perspective of tf and tf-idf, manually select the ontologies to build up a roadmap integrated a three-line defense of supply chain resilience to matches the before, during, after disruption stages.","PeriodicalId":38808,"journal":{"name":"Transactions on Electrical Engineering, Electronics, and Communications","volume":"82 3 Suppl 1 1","pages":"165-170"},"PeriodicalIF":0.0,"publicationDate":"2023-03-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77496612","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 : 2023-03-22DOI: 10.1109/ECTIDAMTNCON57770.2023.10139507
Worrakit Sanpote, Ponnipa Jantawong, Narit Hnoohom, A. Jitpattanakul, S. Mekruksavanich
Nowadays, one of the most important objectives in health-related research is the improvement of the living condition and well-being of people. Smart home systems can provide health protection for residents based on the results of daily activity recognition. Recent advances and developments in sensor technology have increased the need for sensor-compatible goods and services in smart homes. Consequently, the ever-increasing volume of data requires the field of deep learning (DL) for auto-matic human motion recognition. Recent research has modeled spatiotemporal sequences gathered by smart home sensors using long short-term memory networks. In this work, ResNeXt-based models that learn to classify human activities in smart homes were proposed to improve recognition performance. Experiments conducted on Center for Advanced Studies in Adaptive Systems (CASAS) data, a publicly available benchmark dataset, shows that the proposed ResNeXt-based techniques are significantly superior to the existing DL methods and provide better results compared to the existing literature. The ResNeXt model achieved the averaged accuracy over the benchmark method to 84.81%, 93.57%, and 90.38% for the CASAS_Cairo, CASAS_Milan and CASAS_Kyoto3 datasets, respectively.
当今，健康相关研究的重要目标之一是改善人们的生活条件和福祉。智能家居系统可以根据日常活动识别的结果为居民提供健康保护。传感器技术的最新进步和发展增加了智能家居中对传感器兼容商品和服务的需求。因此，不断增加的数据量需要深度学习(DL)领域的自动人体运动识别。最近的研究利用长短期记忆网络对智能家居传感器收集的时空序列进行了建模。在这项工作中，提出了基于resnext的模型来学习对智能家居中的人类活动进行分类，以提高识别性能。在自适应系统高级研究中心(Center for Advanced Studies in Adaptive Systems, CASAS)数据(一个公开的基准数据集)上进行的实验表明，所提出的基于resnext的技术明显优于现有的深度学习方法，并且与现有文献相比提供了更好的结果。对于CASAS_Cairo、CASAS_Milan和CASAS_Kyoto3数据集，ResNeXt模型比基准方法的平均准确率分别达到84.81%、93.57%和90.38%。
{"title":"Deep Learning Approaches for Recognizing Daily Human Activities Using Smart Home Sensors","authors":"Worrakit Sanpote, Ponnipa Jantawong, Narit Hnoohom, A. Jitpattanakul, S. Mekruksavanich","doi":"10.1109/ECTIDAMTNCON57770.2023.10139507","DOIUrl":"https://doi.org/10.1109/ECTIDAMTNCON57770.2023.10139507","url":null,"abstract":"Nowadays, one of the most important objectives in health-related research is the improvement of the living condition and well-being of people. Smart home systems can provide health protection for residents based on the results of daily activity recognition. Recent advances and developments in sensor technology have increased the need for sensor-compatible goods and services in smart homes. Consequently, the ever-increasing volume of data requires the field of deep learning (DL) for auto-matic human motion recognition. Recent research has modeled spatiotemporal sequences gathered by smart home sensors using long short-term memory networks. In this work, ResNeXt-based models that learn to classify human activities in smart homes were proposed to improve recognition performance. Experiments conducted on Center for Advanced Studies in Adaptive Systems (CASAS) data, a publicly available benchmark dataset, shows that the proposed ResNeXt-based techniques are significantly superior to the existing DL methods and provide better results compared to the existing literature. The ResNeXt model achieved the averaged accuracy over the benchmark method to 84.81%, 93.57%, and 90.38% for the CASAS_Cairo, CASAS_Milan and CASAS_Kyoto3 datasets, respectively.","PeriodicalId":38808,"journal":{"name":"Transactions on Electrical Engineering, Electronics, and Communications","volume":"6 1","pages":"469-473"},"PeriodicalIF":0.0,"publicationDate":"2023-03-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78526554","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：