Pub Date : 2021-05-14DOI: 10.1109/DDCLS52934.2021.9455515
Yanzhu Zhang, Haishuai Zhang, Xiaomeng Zhang, J. Pu, Xiaoyan Wang
When light with object information passes through a multi-core fiber, the speckle pattern is obtained. The reconstruction of the original image from the speckle pattern is crucial. In this paper, we propose a lightweight adversarial network for reconstruct image from the speckle pattern. Combining the characteristics of U-Net network and Mobile-Net, a lightweight Mobile-U-Net network is formed to reduce the number of network parameters by using deep separable convolution to realize fast reconstructing image. The adversarial network is also introduced to restrain the quality of the restored image and solve the quality problem of the restored image further. Thus, a high-quality reconstructing image can be achieved.
{"title":"Fiber Optic Speckle Recovery Based on Lightweight Adversarial Network","authors":"Yanzhu Zhang, Haishuai Zhang, Xiaomeng Zhang, J. Pu, Xiaoyan Wang","doi":"10.1109/DDCLS52934.2021.9455515","DOIUrl":"https://doi.org/10.1109/DDCLS52934.2021.9455515","url":null,"abstract":"When light with object information passes through a multi-core fiber, the speckle pattern is obtained. The reconstruction of the original image from the speckle pattern is crucial. In this paper, we propose a lightweight adversarial network for reconstruct image from the speckle pattern. Combining the characteristics of U-Net network and Mobile-Net, a lightweight Mobile-U-Net network is formed to reduce the number of network parameters by using deep separable convolution to realize fast reconstructing image. The adversarial network is also introduced to restrain the quality of the restored image and solve the quality problem of the restored image further. Thus, a high-quality reconstructing image can be achieved.","PeriodicalId":325897,"journal":{"name":"2021 IEEE 10th Data Driven Control and Learning Systems Conference (DDCLS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115229047","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}
Aiming at the non-linearity and uncertainty of patient spastic disturbance in the trajectory tracking control of upper limb rehabilitation robot, an iterative learning control algorithm is proposed based on RBF neural network. This paper considers repetitive nature of the rehabilitation robot system, the algorithm combines a single hidden layer feedforward neural network with iterative learning. In the upper limb rehabilitation process, the algorithm accelerate the convergence speed of the trajectory tracking error, and quickly suppress the interference in the interference environment. The Lyapunov stability theory is used to prove the globally asymptotic stability of the closed-loop system, then simulation proves the feasibility and effectiveness of the proposed algorithm.
{"title":"An Iterative Learning Algorithm Based on RBF Neural Network in Upper Limb Rehabilitation Robot","authors":"Zaixiang Pang, Tongyu Wang, Shuai Liu, Zhanli Wang, Xiyu Zhang, Yan Hao","doi":"10.1109/ddcls52934.2021.9455640","DOIUrl":"https://doi.org/10.1109/ddcls52934.2021.9455640","url":null,"abstract":"Aiming at the non-linearity and uncertainty of patient spastic disturbance in the trajectory tracking control of upper limb rehabilitation robot, an iterative learning control algorithm is proposed based on RBF neural network. This paper considers repetitive nature of the rehabilitation robot system, the algorithm combines a single hidden layer feedforward neural network with iterative learning. In the upper limb rehabilitation process, the algorithm accelerate the convergence speed of the trajectory tracking error, and quickly suppress the interference in the interference environment. The Lyapunov stability theory is used to prove the globally asymptotic stability of the closed-loop system, then simulation proves the feasibility and effectiveness of the proposed algorithm.","PeriodicalId":325897,"journal":{"name":"2021 IEEE 10th Data Driven Control and Learning Systems Conference (DDCLS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115285787","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}
Polyp of intestinal tract is the precursor of colorectal cancer. Accurate computer-aided polyp location and segmentation in colonoscopy is of great importance since it provides valuable information for endoscopists. However, polyps are arduous to be segmented due to their high inter-class similarity, high intra-class variation, and low contrast with surrounding mucosa. To address these challenges, we propose a multi-scale boundary network (MSB-Net) for polyp segmentation. We first focus on the multi-scale feature representation and propose a novel architectural unit to extract intra-stage and contextual information, which is named ResU-Block (RUB). RUBs are connected by the proposed multi-squeeze-and-excitation (Multi-SE) units which can recalibrate the feature information from a multi-scale perspective. We then generate a coarse prediction using the partial decoder, of which the boundary is further refined by a shallow-level attention (SA) module. In addition, we exploit the boundary details using a set of reverse attention (RA) modules, which can progressively establish relationships between regions and boundaries from deep-level features. Comprehensive experiments on five public datasets across five metrics elucidate that our architecture outperforms other SOTA methods by a large margin while maintaining comparable model complexity and inference speed.
{"title":"MSB-Net: Multi-Scale Boundary Net for Polyp Segmentation","authors":"Dongchao Wang, Mingjie Hao, Ruirui Xia, Jinhui Zhu, Sheng Li, Xiongxiong He","doi":"10.1109/DDCLS52934.2021.9455514","DOIUrl":"https://doi.org/10.1109/DDCLS52934.2021.9455514","url":null,"abstract":"Polyp of intestinal tract is the precursor of colorectal cancer. Accurate computer-aided polyp location and segmentation in colonoscopy is of great importance since it provides valuable information for endoscopists. However, polyps are arduous to be segmented due to their high inter-class similarity, high intra-class variation, and low contrast with surrounding mucosa. To address these challenges, we propose a multi-scale boundary network (MSB-Net) for polyp segmentation. We first focus on the multi-scale feature representation and propose a novel architectural unit to extract intra-stage and contextual information, which is named ResU-Block (RUB). RUBs are connected by the proposed multi-squeeze-and-excitation (Multi-SE) units which can recalibrate the feature information from a multi-scale perspective. We then generate a coarse prediction using the partial decoder, of which the boundary is further refined by a shallow-level attention (SA) module. In addition, we exploit the boundary details using a set of reverse attention (RA) modules, which can progressively establish relationships between regions and boundaries from deep-level features. Comprehensive experiments on five public datasets across five metrics elucidate that our architecture outperforms other SOTA methods by a large margin while maintaining comparable model complexity and inference speed.","PeriodicalId":325897,"journal":{"name":"2021 IEEE 10th Data Driven Control and Learning Systems Conference (DDCLS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116949630","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 : 2021-05-14DOI: 10.1109/DDCLS52934.2021.9455535
Kexin Zhang, Yong Liu
This paper proposes an unsupervised feature learning approach on industrial time series data for detection of valve stiction. Considering the commonly existed characteristics of industrial time series signals and the condition that sometimes massive reliable labeled-data are not available, a new time series data transformation and augmentation method is developed. The transformation stage converts the raw time series signals to 2-D matrices and the augmentation stage increases the diversity of the matrices by performing transformation on different timescales. Then a convolutional autoencoder is used to extract the representative features on the augmented data, these new features are taken as the inputs of the traditional clustering algorithms. Unlike the traditional approaches using hand-crafted features or requiring labeled-data, the proposed strategy can automatically learn features on the time series data collected from industrial control loops without supervision. The effectiveness of the proposed approach is evaluated through the International Stiction Data Base (ISDB). Compared with the traditional machine learning methods and deep learning based methods, the experimental results demonstrate that the proposed strategy outperforms the other methods. Besides performance evaluation, we provide a visualization process of feature learning via principal component analysis.
{"title":"Unsupervised Feature Learning with Data Augmentation for Control Valve Stiction Detection","authors":"Kexin Zhang, Yong Liu","doi":"10.1109/DDCLS52934.2021.9455535","DOIUrl":"https://doi.org/10.1109/DDCLS52934.2021.9455535","url":null,"abstract":"This paper proposes an unsupervised feature learning approach on industrial time series data for detection of valve stiction. Considering the commonly existed characteristics of industrial time series signals and the condition that sometimes massive reliable labeled-data are not available, a new time series data transformation and augmentation method is developed. The transformation stage converts the raw time series signals to 2-D matrices and the augmentation stage increases the diversity of the matrices by performing transformation on different timescales. Then a convolutional autoencoder is used to extract the representative features on the augmented data, these new features are taken as the inputs of the traditional clustering algorithms. Unlike the traditional approaches using hand-crafted features or requiring labeled-data, the proposed strategy can automatically learn features on the time series data collected from industrial control loops without supervision. The effectiveness of the proposed approach is evaluated through the International Stiction Data Base (ISDB). Compared with the traditional machine learning methods and deep learning based methods, the experimental results demonstrate that the proposed strategy outperforms the other methods. Besides performance evaluation, we provide a visualization process of feature learning via principal component analysis.","PeriodicalId":325897,"journal":{"name":"2021 IEEE 10th Data Driven Control and Learning Systems Conference (DDCLS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116156175","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 : 2021-05-14DOI: 10.1109/DDCLS52934.2021.9455628
Wei Wang, Wenzhou Lu, Keliang Zhou, Qigao Fan
This paper proposed a fractional-order new generation of $6kpm 1$-order repetitive control scheme (FO-NG-$6kpm 1$ RC). FO-NG-$6kpm 1$ RC is composed of NG-$6kpm 1$ RC and a Farrow structure fractional delay filter based on Taylor series expansion. Since $6 kpm 1$ RC/FO-NG $6kpm 1$ RC occupies less digital memory space, it has better dynamic performance than CRC/FO-CRC, which is more suitable for the control of three-phase power systems dominated by the $6kpm 1$-order harmonics. By updating filter coefficients online in real time, FO-NG-$6kpm 1$ RC can quickly and effectively deal with the frequency variations of grid-connected converters, and has good frequency adaptability. Through a simulation example applied to a three-phase grid-connected inverter, the effectiveness and superiority of the proposed FO-NG-$6kpm 1$ RC control scheme are fully verified.
{"title":"Fractional-Order New Generation of $6k+1$-Order Repetitive Control for Three-phase Grid-connected Converters","authors":"Wei Wang, Wenzhou Lu, Keliang Zhou, Qigao Fan","doi":"10.1109/DDCLS52934.2021.9455628","DOIUrl":"https://doi.org/10.1109/DDCLS52934.2021.9455628","url":null,"abstract":"This paper proposed a fractional-order new generation of <tex>$6kpm 1$</tex>-order repetitive control scheme (FO-NG-<tex>$6kpm 1$</tex> RC). FO-NG-<tex>$6kpm 1$</tex> RC is composed of NG-<tex>$6kpm 1$</tex> RC and a Farrow structure fractional delay filter based on Taylor series expansion. Since <tex>$6 kpm 1$</tex> RC/FO-NG <tex>$6kpm 1$</tex> RC occupies less digital memory space, it has better dynamic performance than CRC/FO-CRC, which is more suitable for the control of three-phase power systems dominated by the <tex>$6kpm 1$</tex>-order harmonics. By updating filter coefficients online in real time, FO-NG-<tex>$6kpm 1$</tex> RC can quickly and effectively deal with the frequency variations of grid-connected converters, and has good frequency adaptability. Through a simulation example applied to a three-phase grid-connected inverter, the effectiveness and superiority of the proposed FO-NG-<tex>$6kpm 1$</tex> RC control scheme are fully verified.","PeriodicalId":325897,"journal":{"name":"2021 IEEE 10th Data Driven Control and Learning Systems Conference (DDCLS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121995009","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 : 2021-05-14DOI: 10.1109/DDCLS52934.2021.9455621
Zhenyu Liu, Yan Li
In this paper, the fault diagnosis scheme for battery is investigated by an improved radial basis function (RBF) neural network. First, the causes of battery faults and the difficulties of fault diagnosis are analyzed. Second, by using the characteristics of experimental data, the subtractive clustering method (SCM) is employed to determine the number of hidden layer neurons, center vector, and expansion coefficient in the RBF neural network. Then, a battery fault diagnosis scheme is designed based on the proposed improved RBF neural network. The simulation results show that the designed scheme can accurately diagnose the type of battery fault with a fast training speed.
{"title":"Battery Fault Diagnosis Scheme Based on Improved RBF Neural Network","authors":"Zhenyu Liu, Yan Li","doi":"10.1109/DDCLS52934.2021.9455621","DOIUrl":"https://doi.org/10.1109/DDCLS52934.2021.9455621","url":null,"abstract":"In this paper, the fault diagnosis scheme for battery is investigated by an improved radial basis function (RBF) neural network. First, the causes of battery faults and the difficulties of fault diagnosis are analyzed. Second, by using the characteristics of experimental data, the subtractive clustering method (SCM) is employed to determine the number of hidden layer neurons, center vector, and expansion coefficient in the RBF neural network. Then, a battery fault diagnosis scheme is designed based on the proposed improved RBF neural network. The simulation results show that the designed scheme can accurately diagnose the type of battery fault with a fast training speed.","PeriodicalId":325897,"journal":{"name":"2021 IEEE 10th Data Driven Control and Learning Systems Conference (DDCLS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129548587","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 : 2021-05-14DOI: 10.1109/DDCLS52934.2021.9455487
Huanlei Guo, Jie Wang, Tingwei Zhou, Wenkang Huang, Junqing Yuan, Xiongxiong He
Single image deraining is regarded as an important research direction in image processing. To tackle the over-smoothing effect caused by the overlapping between rain streaks and the background, we propose a multi-fusion network for single image deraining. A novel local feature fusion block and a global feature fusion block are explored to fuse the high-level features with the low-level ones and correct the low-level representations. By stacking multiple fusion blocks, the proposed network can fully utilize the high-level information and extract powerful feature maps of rain streak layers. In addition, based on the prediction difficulty, a curriculum learning strategy is further explored to make the training process easier. Extensive experiments demonstrate that our network performs favorably against other deraining approaches.
{"title":"Multi-fusion Network for Single Image Deraining","authors":"Huanlei Guo, Jie Wang, Tingwei Zhou, Wenkang Huang, Junqing Yuan, Xiongxiong He","doi":"10.1109/DDCLS52934.2021.9455487","DOIUrl":"https://doi.org/10.1109/DDCLS52934.2021.9455487","url":null,"abstract":"Single image deraining is regarded as an important research direction in image processing. To tackle the over-smoothing effect caused by the overlapping between rain streaks and the background, we propose a multi-fusion network for single image deraining. A novel local feature fusion block and a global feature fusion block are explored to fuse the high-level features with the low-level ones and correct the low-level representations. By stacking multiple fusion blocks, the proposed network can fully utilize the high-level information and extract powerful feature maps of rain streak layers. In addition, based on the prediction difficulty, a curriculum learning strategy is further explored to make the training process easier. Extensive experiments demonstrate that our network performs favorably against other deraining approaches.","PeriodicalId":325897,"journal":{"name":"2021 IEEE 10th Data Driven Control and Learning Systems Conference (DDCLS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128974076","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 : 2021-05-14DOI: 10.1109/DDCLS52934.2021.9455664
Yongshuai Wang, Zengqiang Chen, Mingwei Sun, Qinglin Sun
Load frequency control is an important issue in power systems, so focusing on the typical two-area interconnected power system with non-reheat turbines, this paper designed the learning active disturbance rejection controller to achieve intelligent and adaptive tuning of control parameters, in which the deep reinforcement learning is adopted to adapt to unexpected uncertainties and faults, even a new environment. Finally, numerical simulations show the better performance of the learning controller, and the strong capability to deal with uncertainties and disturbances comparing with the general LADRC controller.
{"title":"Load frequency active disturbance rejection control for an interconnected power system via deep reinforcement learning","authors":"Yongshuai Wang, Zengqiang Chen, Mingwei Sun, Qinglin Sun","doi":"10.1109/DDCLS52934.2021.9455664","DOIUrl":"https://doi.org/10.1109/DDCLS52934.2021.9455664","url":null,"abstract":"Load frequency control is an important issue in power systems, so focusing on the typical two-area interconnected power system with non-reheat turbines, this paper designed the learning active disturbance rejection controller to achieve intelligent and adaptive tuning of control parameters, in which the deep reinforcement learning is adopted to adapt to unexpected uncertainties and faults, even a new environment. Finally, numerical simulations show the better performance of the learning controller, and the strong capability to deal with uncertainties and disturbances comparing with the general LADRC controller.","PeriodicalId":325897,"journal":{"name":"2021 IEEE 10th Data Driven Control and Learning Systems Conference (DDCLS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124177757","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 : 2021-05-14DOI: 10.1109/DDCLS52934.2021.9455667
Meng-Ying Su, Weiwei Che, Zhenling Wang
The model-free adaptive security tracking control (MFASTC) problem of nonlinear networked control systems is explored in this paper with DoS attacks and delays consideration. In order to alleviate the impact of DoS attack and RTT delays on NCSs performance, an attack compensation mechanism and a networked predictive-based delay compensation mechanism are designed, respectively. The data-based designed method need not the dynamic and structure of the system, The MFASTC algorithm is proposed to ensure the output tracking error being bounded in the mean-square sense. Finally, an example is given to illustrate the effectiveness of the new algorithm by a comparison.
{"title":"Model-Free Adaptive Security Tracking Control for Networked Control Systems","authors":"Meng-Ying Su, Weiwei Che, Zhenling Wang","doi":"10.1109/DDCLS52934.2021.9455667","DOIUrl":"https://doi.org/10.1109/DDCLS52934.2021.9455667","url":null,"abstract":"The model-free adaptive security tracking control (MFASTC) problem of nonlinear networked control systems is explored in this paper with DoS attacks and delays consideration. In order to alleviate the impact of DoS attack and RTT delays on NCSs performance, an attack compensation mechanism and a networked predictive-based delay compensation mechanism are designed, respectively. The data-based designed method need not the dynamic and structure of the system, The MFASTC algorithm is proposed to ensure the output tracking error being bounded in the mean-square sense. Finally, an example is given to illustrate the effectiveness of the new algorithm by a comparison.","PeriodicalId":325897,"journal":{"name":"2021 IEEE 10th Data Driven Control and Learning Systems Conference (DDCLS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123371218","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 : 2021-05-14DOI: 10.1109/DDCLS52934.2021.9455547
Yi Xiao, Hongru Ren, Renquan Lu, Shen Cheng
In product life cycle, it is very important to use the manufacturing big data to build prediction model and apply it to predict whether the design task of the product can be completed within the specified time. Most of the existing prediction models in manufacturing industry are built by a single algorithm or its improved version, and neglect the limitation of using a single forecasting algorithm, which may lead to poor forecasting accuracy. This paper aims to integrate the K-nearest neighbor classification algorithm and the logistic regression algorithm linearly in parallel to obtain the combined model which is called K-nearest neighbor-logistic regression (KNN-LR) in this paper, and use the combined model to predict whether the design task of the product can be completed within the specified time. Experimental results show that compared with the model built by a single algorithm, the combined model has better performance on model evaluation indicators such as accuracy, precision, F1 value. recall and classification error rate.
{"title":"Manufacturing Big Data Modeling Based on KNN-LR Algorithm and Its Application in Product Design Business Domain","authors":"Yi Xiao, Hongru Ren, Renquan Lu, Shen Cheng","doi":"10.1109/DDCLS52934.2021.9455547","DOIUrl":"https://doi.org/10.1109/DDCLS52934.2021.9455547","url":null,"abstract":"In product life cycle, it is very important to use the manufacturing big data to build prediction model and apply it to predict whether the design task of the product can be completed within the specified time. Most of the existing prediction models in manufacturing industry are built by a single algorithm or its improved version, and neglect the limitation of using a single forecasting algorithm, which may lead to poor forecasting accuracy. This paper aims to integrate the K-nearest neighbor classification algorithm and the logistic regression algorithm linearly in parallel to obtain the combined model which is called K-nearest neighbor-logistic regression (KNN-LR) in this paper, and use the combined model to predict whether the design task of the product can be completed within the specified time. Experimental results show that compared with the model built by a single algorithm, the combined model has better performance on model evaluation indicators such as accuracy, precision, F1 value. recall and classification error rate.","PeriodicalId":325897,"journal":{"name":"2021 IEEE 10th Data Driven Control and Learning Systems Conference (DDCLS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114878158","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：