Pub Date : 2022-05-01DOI: 10.1109/cniot55862.2022.00025
Wei Luo, Lizhi Zhang, Linbo Xu
Fishing WiFi hotspots are highly concealed and harmful. Traditional positioning algorithms often cannot be directly applied to the tracking and positioning of illegal signal sources due to high cost, difficulty in deployment, and low flexibility. In light of this, we proposed a positioning method of WiFi signal source, which is used in the scene of detecting and tracking fake APs. After collecting the signal data onto the idea of crowdsensing, the coordinates of the centroid of multiple groups is preliminarily calculated by the triangular centroid positioning method, and then the results are processed by the k-means clustering algorithm, and the appropriate weight value is selected according to the size of each cluster, and calculate the final result. The experimental results show that when the transmission path loss factor n=2, the average error of this method is only 34.389% of the triangular centroid location algorithm, and 56.346% of the weighted centroid location method. It not only ensures the accuracy of the calculation results, but also has strong anti-interference ability.
{"title":"An Improved Multi-Centroid Localization Algorithm for WiFi Signal Source Tracking","authors":"Wei Luo, Lizhi Zhang, Linbo Xu","doi":"10.1109/cniot55862.2022.00025","DOIUrl":"https://doi.org/10.1109/cniot55862.2022.00025","url":null,"abstract":"Fishing WiFi hotspots are highly concealed and harmful. Traditional positioning algorithms often cannot be directly applied to the tracking and positioning of illegal signal sources due to high cost, difficulty in deployment, and low flexibility. In light of this, we proposed a positioning method of WiFi signal source, which is used in the scene of detecting and tracking fake APs. After collecting the signal data onto the idea of crowdsensing, the coordinates of the centroid of multiple groups is preliminarily calculated by the triangular centroid positioning method, and then the results are processed by the k-means clustering algorithm, and the appropriate weight value is selected according to the size of each cluster, and calculate the final result. The experimental results show that when the transmission path loss factor n=2, the average error of this method is only 34.389% of the triangular centroid location algorithm, and 56.346% of the weighted centroid location method. It not only ensures the accuracy of the calculation results, but also has strong anti-interference ability.","PeriodicalId":251734,"journal":{"name":"2022 3rd International Conference on Computing, Networks and Internet of Things (CNIOT)","volume":"104 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124208840","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 : 2022-05-01DOI: 10.1109/cniot55862.2022.00024
Yuan Ruan, Qinghua Chen, Xiang-lin Pan
The traffic Internet of Things simulation teaching resource database is an important part of the IoT experimental teaching. This paper constructs a four-tier architecture system of the resource database including hardware level, software level, resource level and application level, analyzes the types, collection technology and storage management methods of simulation teaching resources, and puts forward three resource sharing application modes that are in class and out of class sharing, professional internal and external sharing as well as inside and outside the school sharing, so as to realize the for traffic IoT course teaching mode of combining the theory and practice, both the online and offline available, and no suspension of classes.
{"title":"Research on the Construction of Simulation Teaching Resource Library for Internet of Things in Transportation","authors":"Yuan Ruan, Qinghua Chen, Xiang-lin Pan","doi":"10.1109/cniot55862.2022.00024","DOIUrl":"https://doi.org/10.1109/cniot55862.2022.00024","url":null,"abstract":"The traffic Internet of Things simulation teaching resource database is an important part of the IoT experimental teaching. This paper constructs a four-tier architecture system of the resource database including hardware level, software level, resource level and application level, analyzes the types, collection technology and storage management methods of simulation teaching resources, and puts forward three resource sharing application modes that are in class and out of class sharing, professional internal and external sharing as well as inside and outside the school sharing, so as to realize the for traffic IoT course teaching mode of combining the theory and practice, both the online and offline available, and no suspension of classes.","PeriodicalId":251734,"journal":{"name":"2022 3rd International Conference on Computing, Networks and Internet of Things (CNIOT)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126088984","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 : 2022-05-01DOI: 10.1109/cniot55862.2022.00011
Shuhong Chen, Zhiyi Zhuo
Based on big data analysis, we discuss how to formulate an optimal coping mechanism for infectious diseases, especially major and emerging infectious diseases. First, by combining big data analysis and statistical analysis model and deducing whether the emerging disease is contagious, the strength of the contagion effect and the possible consequences, this study will determine whether the corresponding coping strategies should be implemented for infectious diseases, especially major and emerging infectious diseases. Secondly, according to the inspection results and actual situation, the optimal coping strategy is formulated to minimize the loss of life and property security of the country and the society by using the optimization principle and the objective management in management science. Finally, the statistical analysis method and the six sigma principle are combined to develop a feedback mechanism to evaluate whether the formulated coping strategies can achieve the expected results in practice. Our research has improved the research framework of infectious diseases in theory and provided scientific reference and experience for the major and emerging infectious diseases in practice for the future.
{"title":"New Ideas and Methods of Coping Mechanism for Infectious Diseases Based on Big Data: A Critical Literature Review","authors":"Shuhong Chen, Zhiyi Zhuo","doi":"10.1109/cniot55862.2022.00011","DOIUrl":"https://doi.org/10.1109/cniot55862.2022.00011","url":null,"abstract":"Based on big data analysis, we discuss how to formulate an optimal coping mechanism for infectious diseases, especially major and emerging infectious diseases. First, by combining big data analysis and statistical analysis model and deducing whether the emerging disease is contagious, the strength of the contagion effect and the possible consequences, this study will determine whether the corresponding coping strategies should be implemented for infectious diseases, especially major and emerging infectious diseases. Secondly, according to the inspection results and actual situation, the optimal coping strategy is formulated to minimize the loss of life and property security of the country and the society by using the optimization principle and the objective management in management science. Finally, the statistical analysis method and the six sigma principle are combined to develop a feedback mechanism to evaluate whether the formulated coping strategies can achieve the expected results in practice. Our research has improved the research framework of infectious diseases in theory and provided scientific reference and experience for the major and emerging infectious diseases in practice for the future.","PeriodicalId":251734,"journal":{"name":"2022 3rd International Conference on Computing, Networks and Internet of Things (CNIOT)","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127110829","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 : 2022-05-01DOI: 10.1109/cniot55862.2022.00028
Xiaolong Liu, Ying Pan
In the context of big data, faster and more accurate methods are required for RDF data retrieval. The current research on querying RDF graph data has made some progress, but it has a certain delay and high up-front cost. Given the above shortcomings, we propose a more efficient framework for querying RDF graph data based on the pay-as-you-go (PAYG) approach. Firstly, we annotate the evolution process of data content and association and then construct the evolution update operation set and dynamic incremental graph to describe the dynamic data. Secondly, we design a query algorithm supporting the best-effort query, which returns the data information with the highest similarity to the user, thus improving the search efficiency. Finally, we apply the investment income theory and information retrieval evaluation methods to construct an evaluation mechanism for PAYG RDF data management.
{"title":"A Query Framework for Massive RDF Graph Data in Pay-As-You-Go Fashion","authors":"Xiaolong Liu, Ying Pan","doi":"10.1109/cniot55862.2022.00028","DOIUrl":"https://doi.org/10.1109/cniot55862.2022.00028","url":null,"abstract":"In the context of big data, faster and more accurate methods are required for RDF data retrieval. The current research on querying RDF graph data has made some progress, but it has a certain delay and high up-front cost. Given the above shortcomings, we propose a more efficient framework for querying RDF graph data based on the pay-as-you-go (PAYG) approach. Firstly, we annotate the evolution process of data content and association and then construct the evolution update operation set and dynamic incremental graph to describe the dynamic data. Secondly, we design a query algorithm supporting the best-effort query, which returns the data information with the highest similarity to the user, thus improving the search efficiency. Finally, we apply the investment income theory and information retrieval evaluation methods to construct an evaluation mechanism for PAYG RDF data management.","PeriodicalId":251734,"journal":{"name":"2022 3rd International Conference on Computing, Networks and Internet of Things (CNIOT)","volume":"35 4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123189202","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 : 2022-05-01DOI: 10.1109/cniot55862.2022.00032
Xinjie Tan, Zhidong Xie, Xinwang Yuan, Gang Yang, Yung-Su Han
With the development of wireless communication technology and the updates of communication equipment, the modulation of signal becomes more complex, and modulation recognition is becoming more and more difficult. Traditional signal modulation recognition methods rely on human experience, its feature extraction process is complex, and the empirical threshold is difficult to find. The recognition method combined with manual feature extraction and deep neural network can achieve better recognition accuracy, but it is still limited by the process of feature extraction. Compared with the above, automatic modulation recognition method based on deep learning is more efficient in complicated open environment. In this paper, a residual neural network for automatic modulation recognition was designed, and the experiment had achieved remarkable results. When SNR is 10dB, we got an accuracy of 95.3% faced to 24 kinds of signals, and when SNR is 12dB, we got an accuracy of 96.3%. Compared with existing models, this model reduces the network parameters, greatly shortens the training time, and lower the hardware requirements. This model shows a good result on the recognition of high-level modulation signal. When SNR is 10dB, the recognition accuracy of 128APSK, 128QAM and 256QAM is 97%, 88% and 88%.
{"title":"A Residual Neural Network for Modulation Recognition of 24 kinds of Signals","authors":"Xinjie Tan, Zhidong Xie, Xinwang Yuan, Gang Yang, Yung-Su Han","doi":"10.1109/cniot55862.2022.00032","DOIUrl":"https://doi.org/10.1109/cniot55862.2022.00032","url":null,"abstract":"With the development of wireless communication technology and the updates of communication equipment, the modulation of signal becomes more complex, and modulation recognition is becoming more and more difficult. Traditional signal modulation recognition methods rely on human experience, its feature extraction process is complex, and the empirical threshold is difficult to find. The recognition method combined with manual feature extraction and deep neural network can achieve better recognition accuracy, but it is still limited by the process of feature extraction. Compared with the above, automatic modulation recognition method based on deep learning is more efficient in complicated open environment. In this paper, a residual neural network for automatic modulation recognition was designed, and the experiment had achieved remarkable results. When SNR is 10dB, we got an accuracy of 95.3% faced to 24 kinds of signals, and when SNR is 12dB, we got an accuracy of 96.3%. Compared with existing models, this model reduces the network parameters, greatly shortens the training time, and lower the hardware requirements. This model shows a good result on the recognition of high-level modulation signal. When SNR is 10dB, the recognition accuracy of 128APSK, 128QAM and 256QAM is 97%, 88% and 88%.","PeriodicalId":251734,"journal":{"name":"2022 3rd International Conference on Computing, Networks and Internet of Things (CNIOT)","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115089814","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 : 2022-05-01DOI: 10.1109/cniot55862.2022.00021
Chenchen Ji, Yanjun Wu, Pengpeng Hou, Yang Tai, Jiageng Yu
The cloud-only and edge-computing approaches have recently been proposed to satisfy the requirements of complex neural networks. However, the cloud-only approach generates a latency challenge because of the high data volumes that must be sent to a centralized location in the cloud. Less-powerful edge computing resources require a compression model for computation reduction, which degrades the model trading accuracy. To address this challenge, deep neural network (DNN) partitioning has become a recent trend, with DNN models being sliced into head and tail portions executed at the mobile edge devices and cloud server, respectively. We propose Edgepipe, a novel partitioning method based on pipeline inference with an image stream to automatically partition DNN computation between the edge device and cloud server, thereby reducing the global latency and enhancing the system-wide real-time performance. This method adapts to various DNN architectures, hardware platforms, and networks. Here, when evaluated on a suite of five DNN applications, Edgepipe achieves average latency speedups of 1.241× and 1.154× over the cloud-only approach and the state-of-the-art approach known as “Neurosurgeon”, respectively.
{"title":"Novel Adaptive DNN Partitioning Method Based on Image-Stream Pipeline Inference between the Edge and Cloud","authors":"Chenchen Ji, Yanjun Wu, Pengpeng Hou, Yang Tai, Jiageng Yu","doi":"10.1109/cniot55862.2022.00021","DOIUrl":"https://doi.org/10.1109/cniot55862.2022.00021","url":null,"abstract":"The cloud-only and edge-computing approaches have recently been proposed to satisfy the requirements of complex neural networks. However, the cloud-only approach generates a latency challenge because of the high data volumes that must be sent to a centralized location in the cloud. Less-powerful edge computing resources require a compression model for computation reduction, which degrades the model trading accuracy. To address this challenge, deep neural network (DNN) partitioning has become a recent trend, with DNN models being sliced into head and tail portions executed at the mobile edge devices and cloud server, respectively. We propose Edgepipe, a novel partitioning method based on pipeline inference with an image stream to automatically partition DNN computation between the edge device and cloud server, thereby reducing the global latency and enhancing the system-wide real-time performance. This method adapts to various DNN architectures, hardware platforms, and networks. Here, when evaluated on a suite of five DNN applications, Edgepipe achieves average latency speedups of 1.241× and 1.154× over the cloud-only approach and the state-of-the-art approach known as “Neurosurgeon”, respectively.","PeriodicalId":251734,"journal":{"name":"2022 3rd International Conference on Computing, Networks and Internet of Things (CNIOT)","volume":"134 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116624910","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 : 2022-05-01DOI: 10.1109/cniot55862.2022.00019
Bo Li, Wei-dong Jin, Junxiao Ren
In recent years, computer vision has been greatly developed in the detection of catenary equipment. With its high efficiency and accuracy, it meets the needs of safety detection of catenary equipment in the safe operation of trains. In the catenary monitoring image, some equipment targets are small, which makes it difficult to identify. To solve this problem, this paper proposes an improved small target detection algorithm -DA-YOLOv4. In this method, Dual Attention Network for Scene Segmentation ( DANet ) is integrated into YOLOv4 model. Position Attention Module ( PAM ) and Channel Attention Module ( CAM ) are applied to enhance the attention of feature extraction network to small targets from two aspects of spatial location and feature channel. The context information is fully utilized to solve the problems of difficult feature extraction and low recognition rate of small targets. Experiments show that the DA-YOLOv4 algorithm can effectively improve the detection effect of small targets in the catenary, and the average detection accuracy on the catenary data set is 77.6 %, which is 4.7 % higher than that of the YOLOv4 network.
近年来，计算机视觉在接触网设备检测方面有了很大的发展。该方法效率高、精度高，满足了列车安全运行中接触网设备安全检测的需要。在接触网监测图像中，一些设备目标较小，给识别带来困难。为了解决这一问题，本文提出了一种改进的小目标检测算法-DA-YOLOv4。该方法将场景分割双注意网络(Dual Attention Network for Scene Segmentation, DANet)集成到YOLOv4模型中。利用位置注意模块(PAM)和通道注意模块(CAM)从空间位置和特征通道两个方面增强特征提取网络对小目标的注意。充分利用上下文信息，解决了小目标特征提取困难、识别率低的问题。实验表明，DA-YOLOv4算法能有效提高接触网中小目标的检测效果，在接触网数据集上的平均检测准确率为77.6%，比YOLOv4网络提高了4.7%。
{"title":"Research on Small Target Detection Algorithm of Catenary Based on DA-YOLOv4","authors":"Bo Li, Wei-dong Jin, Junxiao Ren","doi":"10.1109/cniot55862.2022.00019","DOIUrl":"https://doi.org/10.1109/cniot55862.2022.00019","url":null,"abstract":"In recent years, computer vision has been greatly developed in the detection of catenary equipment. With its high efficiency and accuracy, it meets the needs of safety detection of catenary equipment in the safe operation of trains. In the catenary monitoring image, some equipment targets are small, which makes it difficult to identify. To solve this problem, this paper proposes an improved small target detection algorithm -DA-YOLOv4. In this method, Dual Attention Network for Scene Segmentation ( DANet ) is integrated into YOLOv4 model. Position Attention Module ( PAM ) and Channel Attention Module ( CAM ) are applied to enhance the attention of feature extraction network to small targets from two aspects of spatial location and feature channel. The context information is fully utilized to solve the problems of difficult feature extraction and low recognition rate of small targets. Experiments show that the DA-YOLOv4 algorithm can effectively improve the detection effect of small targets in the catenary, and the average detection accuracy on the catenary data set is 77.6 %, which is 4.7 % higher than that of the YOLOv4 network.","PeriodicalId":251734,"journal":{"name":"2022 3rd International Conference on Computing, Networks and Internet of Things (CNIOT)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121918004","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 : 2022-05-01DOI: 10.1109/cniot55862.2022.00036
Rongcheng Duan, Yao Qin, Haokun He, Chang Cai
The purpose of aspect-level sentiment classification is to predict the sentiment polarity of specific aspect words in a sentence. Recently many works exploit LSTM models based on the attention mechanism. However, the prior work only attends to using the aspect terms to capture the aspect-specific sentiment information in the text. It may cause the mismatch of sentiment when the aspect words are extracted incorrectly. To solve this problem, we propose a simple but effective framework called the Constrained Interaction Network(CIN), which consists of the context-aspect level interaction layer(CAI-Layer), the long and short-term memory network layer(LSTM-Layer), and Constraint Attention layer(CA-Layer). CIN can extract the sentiment features of specific aspects with the assistance of LSTM-Layer and CAI-Layer, which greatly share the attention layer. The experiment conducted on three widely used data sets in SemEval 2014 and Twitter shows that the constrained attention mechanism is always better than other existing attention mechanisms, which also confirms that the CA- Layer can indeed help LSTM to extract the specified aspect-level sentiment characteristics.
{"title":"The Constrained Interaction Network for Aspect-level Sentiment Classification Task","authors":"Rongcheng Duan, Yao Qin, Haokun He, Chang Cai","doi":"10.1109/cniot55862.2022.00036","DOIUrl":"https://doi.org/10.1109/cniot55862.2022.00036","url":null,"abstract":"The purpose of aspect-level sentiment classification is to predict the sentiment polarity of specific aspect words in a sentence. Recently many works exploit LSTM models based on the attention mechanism. However, the prior work only attends to using the aspect terms to capture the aspect-specific sentiment information in the text. It may cause the mismatch of sentiment when the aspect words are extracted incorrectly. To solve this problem, we propose a simple but effective framework called the Constrained Interaction Network(CIN), which consists of the context-aspect level interaction layer(CAI-Layer), the long and short-term memory network layer(LSTM-Layer), and Constraint Attention layer(CA-Layer). CIN can extract the sentiment features of specific aspects with the assistance of LSTM-Layer and CAI-Layer, which greatly share the attention layer. The experiment conducted on three widely used data sets in SemEval 2014 and Twitter shows that the constrained attention mechanism is always better than other existing attention mechanisms, which also confirms that the CA- Layer can indeed help LSTM to extract the specified aspect-level sentiment characteristics.","PeriodicalId":251734,"journal":{"name":"2022 3rd International Conference on Computing, Networks and Internet of Things (CNIOT)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129245449","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}
In this process of active rehabilitation assisted by hand rehabilitation robot, the patient’s hand motion intention, that is, the patient’s gesture recognition, plays an important role. Gesture recognition based on sEMG signal is a hot research topic. Due to the spatial correlation and time non-stationary of sEMG signal, this research topic has many difficulties. In order to solve this problem, we come up with a gesture recognition network GTGR-Net based on sEMG signal, which uses the combination of graph attention network and time convolution network to extract the spatiotemporal information of sEMG signal. We verify the effect of our algorithm on three public data sets and achieve good results, which is better than the other ways.
{"title":"GTGR-Net: Graph Attentional-Temporal Network for Surface-Electromyography-Based Gesture Recognition","authors":"Xiaoxu Jia, Hongbo Wang, Jingjing Luo, Zhiping Lai, Xueze Zhang, Weiqi Zhang, Xiuhong Tang","doi":"10.1109/cniot55862.2022.00039","DOIUrl":"https://doi.org/10.1109/cniot55862.2022.00039","url":null,"abstract":"In this process of active rehabilitation assisted by hand rehabilitation robot, the patient’s hand motion intention, that is, the patient’s gesture recognition, plays an important role. Gesture recognition based on sEMG signal is a hot research topic. Due to the spatial correlation and time non-stationary of sEMG signal, this research topic has many difficulties. In order to solve this problem, we come up with a gesture recognition network GTGR-Net based on sEMG signal, which uses the combination of graph attention network and time convolution network to extract the spatiotemporal information of sEMG signal. We verify the effect of our algorithm on three public data sets and achieve good results, which is better than the other ways.","PeriodicalId":251734,"journal":{"name":"2022 3rd International Conference on Computing, Networks and Internet of Things (CNIOT)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117335384","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 : 2022-05-01DOI: 10.1109/cniot55862.2022.00040
Clarissa Faye G. Gamboa, Matthew B. Concepcion, Antolin J. Alipio, Dan Michael A. Cortez, Andrew G. Bitancor, M. S. Santos, F. A. L. Atienza, M. A. S. Mercado
Companies receive millions of tickets from their clients. Unfortunately, manual ticket routing takes time and relies heavily on human resources. To help automate the ticket routing, text classification can assist as it is the process of categorizing a document into a predetermined class based on its content. One algorithm is the K-Nearest Neighbors (KNN) which is a popular supervised technique but ranks average to lowest compared to other classification models. An improved KNN algorithm utilized clustering and improved the accuracy of the classifier. This paper proposed a further enhancement of this algorithm by adding preprocessing techniques, changing the distance formula, and computing for the k-value rather than choosing one. Two datasets of IT support tickets were used to train and test the algorithms. Results showed that this further enhanced algorithm significantly performed better than the initial algorithm with the highest accuracy score of 97.83% in one dataset while the initial algorithm performed best with an accuracy score of 86.34% using a k-value of 4 in another dataset.
{"title":"Further Enhancement of KNN Algorithm Based on Clustering Applied to IT Support Ticket Routing","authors":"Clarissa Faye G. Gamboa, Matthew B. Concepcion, Antolin J. Alipio, Dan Michael A. Cortez, Andrew G. Bitancor, M. S. Santos, F. A. L. Atienza, M. A. S. Mercado","doi":"10.1109/cniot55862.2022.00040","DOIUrl":"https://doi.org/10.1109/cniot55862.2022.00040","url":null,"abstract":"Companies receive millions of tickets from their clients. Unfortunately, manual ticket routing takes time and relies heavily on human resources. To help automate the ticket routing, text classification can assist as it is the process of categorizing a document into a predetermined class based on its content. One algorithm is the K-Nearest Neighbors (KNN) which is a popular supervised technique but ranks average to lowest compared to other classification models. An improved KNN algorithm utilized clustering and improved the accuracy of the classifier. This paper proposed a further enhancement of this algorithm by adding preprocessing techniques, changing the distance formula, and computing for the k-value rather than choosing one. Two datasets of IT support tickets were used to train and test the algorithms. Results showed that this further enhanced algorithm significantly performed better than the initial algorithm with the highest accuracy score of 97.83% in one dataset while the initial algorithm performed best with an accuracy score of 86.34% using a k-value of 4 in another dataset.","PeriodicalId":251734,"journal":{"name":"2022 3rd International Conference on Computing, Networks and Internet of Things (CNIOT)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133972324","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：