For the Unmanned Combat Aerial Vehicle(UCAV)maneuvering decision in close air combat, the design of reinforcement learning(RL) reward function and the selection of hyperparameters are studied based on the deep Q network algorithm. Considering the angle, range, altitude, and speed factors, an auxiliary reward function is proposed to solve the sparse reward problem of RL. Meanwhile, aiming at the issue of hyperparameter selection in RL, the influence of learning rate, the number of network nodes, and layers on the decision-making system is explored, and a suitable range of parameters is given, which provides a reference for the subsequent research on parameter selection. In addition, the simulation results show that the trained agent can obtain the optimal maneuver strategy in different air combat situations, but it is sensitive to RL hyperparameters.
{"title":"Research on Intelligent Maneuvering Decision in Close Air Combat Based on Deep Q Network","authors":"Tingyu Zhang, Chen Zheng, Mingwei Sun, Yongshuai Wang, Zengqiang Chen","doi":"10.1109/DDCLS58216.2023.10166948","DOIUrl":"https://doi.org/10.1109/DDCLS58216.2023.10166948","url":null,"abstract":"For the Unmanned Combat Aerial Vehicle(UCAV)maneuvering decision in close air combat, the design of reinforcement learning(RL) reward function and the selection of hyperparameters are studied based on the deep Q network algorithm. Considering the angle, range, altitude, and speed factors, an auxiliary reward function is proposed to solve the sparse reward problem of RL. Meanwhile, aiming at the issue of hyperparameter selection in RL, the influence of learning rate, the number of network nodes, and layers on the decision-making system is explored, and a suitable range of parameters is given, which provides a reference for the subsequent research on parameter selection. In addition, the simulation results show that the trained agent can obtain the optimal maneuver strategy in different air combat situations, but it is sensitive to RL hyperparameters.","PeriodicalId":415532,"journal":{"name":"2023 IEEE 12th Data Driven Control and Learning Systems Conference (DDCLS)","volume":"70 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129124372","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 : 2023-05-12DOI: 10.1109/DDCLS58216.2023.10165912
Li Yifan, Cao Rongmin, Hou Zhongsheng, Zhou Hui Xing, Chang Debiao, Jia Jihui
Because the linear ultrasonic motor system has obvious nonlinearity and time-varying. In the operation process, the tracking error, mechanical delays, and other factors will greatly impact the position tracking accuracy. To reduce the linear ultrasonic motor position steady-state error. Sliding mode control (SMC) is invariant to system disturbance and model-free adaptive predictive control (MFAPC) can realize adaptive control only by input and output data of a controlled system, this paper designed a model-free adaptive sliding mode predictive controller (MFASMPC) and proved its stability and convergence Finally, the position control of linear ultrasonic motor based on model-free adaptive sliding mode predictive control method is simulated and analyzed. Theoretical proof and simulation results show that such an algorithm can effectively reduce the steady-state error to meet the control accuracy requirements.
{"title":"Model - free Adaptive Sliding Mode Predictive Control of Linear Ultrasonic Motor","authors":"Li Yifan, Cao Rongmin, Hou Zhongsheng, Zhou Hui Xing, Chang Debiao, Jia Jihui","doi":"10.1109/DDCLS58216.2023.10165912","DOIUrl":"https://doi.org/10.1109/DDCLS58216.2023.10165912","url":null,"abstract":"Because the linear ultrasonic motor system has obvious nonlinearity and time-varying. In the operation process, the tracking error, mechanical delays, and other factors will greatly impact the position tracking accuracy. To reduce the linear ultrasonic motor position steady-state error. Sliding mode control (SMC) is invariant to system disturbance and model-free adaptive predictive control (MFAPC) can realize adaptive control only by input and output data of a controlled system, this paper designed a model-free adaptive sliding mode predictive controller (MFASMPC) and proved its stability and convergence Finally, the position control of linear ultrasonic motor based on model-free adaptive sliding mode predictive control method is simulated and analyzed. Theoretical proof and simulation results show that such an algorithm can effectively reduce the steady-state error to meet the control accuracy requirements.","PeriodicalId":415532,"journal":{"name":"2023 IEEE 12th Data Driven Control and Learning Systems Conference (DDCLS)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123873878","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 : 2023-05-12DOI: 10.1109/DDCLS58216.2023.10166491
Daobo Sun, H. Ji
Event-based object detection is a challenging but promising task, as the nature of sparsity and asynchrony of events is incompatible with state-of-the-art object detection approaches. Conventional deep neural networks do not take advantage of the event camera's high event sampling rate, low power consumption and robustness of brightness changes. Recent works addresses the problem of redundant computations by using a graph representation to model the feature of event streams that the graph representation and graph neural networks for event streams can efficiently extract the meaningful information and reduce the computational complexity. Nevertheless, there is still room for improvement in terms of accuracy and computation efficiency. In this work, we propose a graph-based architecture and a new mechanism for updating the graph, which significantly increases the capacity of graph neural networks while maintaining highly efficient per-event processing. In object detection task, our model achieves higher accuracy and lower FLOPS per event compared to various synchronous/asynchronous methods. To our belief, the framework we proposed is effective and robust, as well as being a significant reduction in the amount of redundant computation.
{"title":"Event-Based Object Detection using Graph Neural Networks","authors":"Daobo Sun, H. Ji","doi":"10.1109/DDCLS58216.2023.10166491","DOIUrl":"https://doi.org/10.1109/DDCLS58216.2023.10166491","url":null,"abstract":"Event-based object detection is a challenging but promising task, as the nature of sparsity and asynchrony of events is incompatible with state-of-the-art object detection approaches. Conventional deep neural networks do not take advantage of the event camera's high event sampling rate, low power consumption and robustness of brightness changes. Recent works addresses the problem of redundant computations by using a graph representation to model the feature of event streams that the graph representation and graph neural networks for event streams can efficiently extract the meaningful information and reduce the computational complexity. Nevertheless, there is still room for improvement in terms of accuracy and computation efficiency. In this work, we propose a graph-based architecture and a new mechanism for updating the graph, which significantly increases the capacity of graph neural networks while maintaining highly efficient per-event processing. In object detection task, our model achieves higher accuracy and lower FLOPS per event compared to various synchronous/asynchronous methods. To our belief, the framework we proposed is effective and robust, as well as being a significant reduction in the amount of redundant computation.","PeriodicalId":415532,"journal":{"name":"2023 IEEE 12th Data Driven Control and Learning Systems Conference (DDCLS)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123908405","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 : 2023-05-12DOI: 10.1109/DDCLS58216.2023.10166530
Liduo Nie, Xin Wang
This study examines the leader-follower consistency issue in a particular class of multiagent systems and provides an event-triggered adaptive control approach. The event-triggered mechanism designed in this paper dramatically reduces the communication load and data transmission, which can better serve practical production applications. It is shown that the suggested control method prevents Zeno behavior and ensures that all signals in a closed-loop system are bounded. The effectiveness of the suggested control method is confirmed by the simulation results.
{"title":"Event-Triggered Adaptive Cooperative Control for Nonstrict-Feedback Nonlinear Multiagent Systems","authors":"Liduo Nie, Xin Wang","doi":"10.1109/DDCLS58216.2023.10166530","DOIUrl":"https://doi.org/10.1109/DDCLS58216.2023.10166530","url":null,"abstract":"This study examines the leader-follower consistency issue in a particular class of multiagent systems and provides an event-triggered adaptive control approach. The event-triggered mechanism designed in this paper dramatically reduces the communication load and data transmission, which can better serve practical production applications. It is shown that the suggested control method prevents Zeno behavior and ensures that all signals in a closed-loop system are bounded. The effectiveness of the suggested control method is confirmed by the simulation results.","PeriodicalId":415532,"journal":{"name":"2023 IEEE 12th Data Driven Control and Learning Systems Conference (DDCLS)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114560780","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 : 2023-05-12DOI: 10.1109/DDCLS58216.2023.10167266
Xueming Song, Ke Zhu, Yuxing Zhao, Jianming Zhang
Automated Guided vehicles (AGVs) provide a better solution to hospital logistics. In this paper, a mathematical model for point-to-point pickup and delivery tasks in a hospital with time windows and capacity constraints based on heterogeneous AGVs fleet is established, and a meta-heuristic algorithm based on ALNS is designed to solve the static scheduling problem of AGVs in the hospital environment. The effectiveness of the proposed algorithm is verified by numerical experiments and comparison with the basic algorithm. Finally, we summarized the direction of the further work.
{"title":"Heterogeneous AGVs Scheduling in Hospital Using ALNS-based Metaheuristic Algorithm","authors":"Xueming Song, Ke Zhu, Yuxing Zhao, Jianming Zhang","doi":"10.1109/DDCLS58216.2023.10167266","DOIUrl":"https://doi.org/10.1109/DDCLS58216.2023.10167266","url":null,"abstract":"Automated Guided vehicles (AGVs) provide a better solution to hospital logistics. In this paper, a mathematical model for point-to-point pickup and delivery tasks in a hospital with time windows and capacity constraints based on heterogeneous AGVs fleet is established, and a meta-heuristic algorithm based on ALNS is designed to solve the static scheduling problem of AGVs in the hospital environment. The effectiveness of the proposed algorithm is verified by numerical experiments and comparison with the basic algorithm. Finally, we summarized the direction of the further work.","PeriodicalId":415532,"journal":{"name":"2023 IEEE 12th Data Driven Control and Learning Systems Conference (DDCLS)","volume":"98 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116158295","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 develops a novel method named wavelet denoising convolutional neural network (WDECNN) for fault diagnosis with background noise. The continuous wavelet transform (CWT) is first applied to transform the measured raw vibration data into time-frequency images which serve as the inputs of WDECNN. Then, a light-weight two-dimensional CNN (2DCNN) model is incorporated in WDECNN to simplify the network architecture, while a wavelet denoising module is also applied in it to achieve high accuracy of fault identification in the noisy environment. Particularly, the wavelet denoising module which consists of wavelet decomposition and denoising is parallel to the 2DCNN model, and the denoising results are integrated into pooling layers in the 2DCNN model. Thus, the denoised information is added to the 2DCNN model to improve its feature learning ability. Finally, the effectiveness of the developed method is validated on Paderborn bearing dataset, which illustrates its fault diagnosis capability under background noise.
{"title":"A Fault Diagnosis Method Based on Wavelet Denoising and 2DCNN under Background Noise","authors":"Kexin Liu, Zhe Li, Wenbin He, Jia Peng, Xudong Wang, Yaonan Wang","doi":"10.1109/DDCLS58216.2023.10167183","DOIUrl":"https://doi.org/10.1109/DDCLS58216.2023.10167183","url":null,"abstract":"This paper develops a novel method named wavelet denoising convolutional neural network (WDECNN) for fault diagnosis with background noise. The continuous wavelet transform (CWT) is first applied to transform the measured raw vibration data into time-frequency images which serve as the inputs of WDECNN. Then, a light-weight two-dimensional CNN (2DCNN) model is incorporated in WDECNN to simplify the network architecture, while a wavelet denoising module is also applied in it to achieve high accuracy of fault identification in the noisy environment. Particularly, the wavelet denoising module which consists of wavelet decomposition and denoising is parallel to the 2DCNN model, and the denoising results are integrated into pooling layers in the 2DCNN model. Thus, the denoised information is added to the 2DCNN model to improve its feature learning ability. Finally, the effectiveness of the developed method is validated on Paderborn bearing dataset, which illustrates its fault diagnosis capability under background noise.","PeriodicalId":415532,"journal":{"name":"2023 IEEE 12th Data Driven Control and Learning Systems Conference (DDCLS)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116268553","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 : 2023-05-12DOI: 10.1109/DDCLS58216.2023.10166437
Chuangkai Zheng, Liuming Zhou, Feng Li
In this paper, the predictive control problem of two-dimensional iterative learning model based on just-in-time learning (JITL) model is studied for batch processes. A new error compensation strategy is proposed based on two-dimensional JITL model by using MPC-ILC integrated control method. Batch axis and time axis are integrated into a comprehensive objective function, and the JITL model is used to solve the problem of large computation of comprehensive objective function. The proposed control algorithm is applied to a typical batch reactor, and the results show that the proposed control strategy has good control performance.
{"title":"Two-Dimensional Model Predictive Iterative Learning Control based on Just-in-Time Learning Method for Batch Processes","authors":"Chuangkai Zheng, Liuming Zhou, Feng Li","doi":"10.1109/DDCLS58216.2023.10166437","DOIUrl":"https://doi.org/10.1109/DDCLS58216.2023.10166437","url":null,"abstract":"In this paper, the predictive control problem of two-dimensional iterative learning model based on just-in-time learning (JITL) model is studied for batch processes. A new error compensation strategy is proposed based on two-dimensional JITL model by using MPC-ILC integrated control method. Batch axis and time axis are integrated into a comprehensive objective function, and the JITL model is used to solve the problem of large computation of comprehensive objective function. The proposed control algorithm is applied to a typical batch reactor, and the results show that the proposed control strategy has good control performance.","PeriodicalId":415532,"journal":{"name":"2023 IEEE 12th Data Driven Control and Learning Systems Conference (DDCLS)","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121516486","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 : 2023-05-12DOI: 10.1109/DDCLS58216.2023.10166447
Yanyan Zhang, Kai Zhang, Pengcheng Yang, Kai-xiang Peng
Due to the difficulty in strip crown prediction caused by multivariable, nonlinear and strong coupling in the hot strip rolling mill (HSRM) process, this paper proposes a strip crown prediction model based on support vector regression (SVR), and uses sparrow search algorithm (SSA) to optimize the parameter C and $sigma$ of the model, so as to improve the generalization ability of the prediction model. The overall performance of the model is evaluated by mean square error (MSE), mean absolute error (MAE), mean absolute percentage error (MAPE) and correlation coefficient $(R^{2})$. It shows that the prediction accuracy and generalization ability of the proposed model are better than the traditional methods. The proposed SSA-SVR model in this paper is successfully applied to the crown prediction of the 2150 production line of Ansteel company. The performance shows that the method can be efficient to predict the steel crown in a real HSRM process.
{"title":"Data Driven Strip Crown Prediction for a Hot Strip Rolling Mill Process","authors":"Yanyan Zhang, Kai Zhang, Pengcheng Yang, Kai-xiang Peng","doi":"10.1109/DDCLS58216.2023.10166447","DOIUrl":"https://doi.org/10.1109/DDCLS58216.2023.10166447","url":null,"abstract":"Due to the difficulty in strip crown prediction caused by multivariable, nonlinear and strong coupling in the hot strip rolling mill (HSRM) process, this paper proposes a strip crown prediction model based on support vector regression (SVR), and uses sparrow search algorithm (SSA) to optimize the parameter C and $sigma$ of the model, so as to improve the generalization ability of the prediction model. The overall performance of the model is evaluated by mean square error (MSE), mean absolute error (MAE), mean absolute percentage error (MAPE) and correlation coefficient $(R^{2})$. It shows that the prediction accuracy and generalization ability of the proposed model are better than the traditional methods. The proposed SSA-SVR model in this paper is successfully applied to the crown prediction of the 2150 production line of Ansteel company. The performance shows that the method can be efficient to predict the steel crown in a real HSRM process.","PeriodicalId":415532,"journal":{"name":"2023 IEEE 12th Data Driven Control and Learning Systems Conference (DDCLS)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122240743","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 : 2023-05-12DOI: 10.1109/DDCLS58216.2023.10166679
Qun Zhu, Qianchuan Zhao, Yuan Xu, Yanlin He
In order to optimize complex industrial processes, an accurate model is essential. The mainstream approach for complex industrial modeling is data-driven soft sensors. However, the accuracy of the established models is often low due to an insufficient amount of effective data, so the method of generating virtual samples has been proposed to achieve data augmentation, but the previous virtual sample generation methods have ignored the correlation between samples. To solve this problem, an effective virtual sample generation method based on Gibbs Sampling algorithm (GS- VSG) is proposed in this paper. In the proposed method, virtual input samples are first generated using the prior knowledge of the original data through the Gibbs Sampling method. Next, a generalized regression neural network (GRNN) model is constructed from the raw data, which is used to predict the output values of the virtual samples. Finally, the input and output parts of the virtual samples are combined to create a virtual sample set, which completes the extension of the original data set. To demonstrate the feasibility of the proposed GS- VSG method, numerical example and real industrial process dataset are used for simulation experiments. The results show that GS- VSG generated samples can improve the model accuracy and is a good technique for virtual sample generation.
{"title":"Novel virtual sample generation using Gibbs Sampling integrated with GRNN for handling small data in soft sensing","authors":"Qun Zhu, Qianchuan Zhao, Yuan Xu, Yanlin He","doi":"10.1109/DDCLS58216.2023.10166679","DOIUrl":"https://doi.org/10.1109/DDCLS58216.2023.10166679","url":null,"abstract":"In order to optimize complex industrial processes, an accurate model is essential. The mainstream approach for complex industrial modeling is data-driven soft sensors. However, the accuracy of the established models is often low due to an insufficient amount of effective data, so the method of generating virtual samples has been proposed to achieve data augmentation, but the previous virtual sample generation methods have ignored the correlation between samples. To solve this problem, an effective virtual sample generation method based on Gibbs Sampling algorithm (GS- VSG) is proposed in this paper. In the proposed method, virtual input samples are first generated using the prior knowledge of the original data through the Gibbs Sampling method. Next, a generalized regression neural network (GRNN) model is constructed from the raw data, which is used to predict the output values of the virtual samples. Finally, the input and output parts of the virtual samples are combined to create a virtual sample set, which completes the extension of the original data set. To demonstrate the feasibility of the proposed GS- VSG method, numerical example and real industrial process dataset are used for simulation experiments. The results show that GS- VSG generated samples can improve the model accuracy and is a good technique for virtual sample generation.","PeriodicalId":415532,"journal":{"name":"2023 IEEE 12th Data Driven Control and Learning Systems Conference (DDCLS)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130002956","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 : 2023-05-12DOI: 10.1109/DDCLS58216.2023.10166293
Cao Yu-kun, Wei Zi-yue, Tang Yi-jia, Jin Cheng-kun
Hierarchical label text classification is a challenging task in the field of natural language processing, where each document needs to be correctly classified into multiple labels with hierarchical structure. However, in the label set, due to the insufficient semantic information contained in the labels and the small number of documents classified under deep-level labels, the training of deep-level labels is insufficient, leading to a significant imbalance in label training. To address this, a hierarchical label text classification method with deep-level label-assisted classification (DLAC) is proposed. The method proposes a deep-level label-assisted classifier, which effectively utilizes text features and rich features of shallow label nodes corresponding to deep label nodes (i.e., shallow label's rich features) on the basis of enhanced label semantics to enhance the classification performance of deep labels. The comparison experiment results with eleven algorithms on three datasets show that the model can effectively improve the classification performance of deep-level labels and achieve good results.
{"title":"Hierarchical Label Text Classification Method with Deep-Level Label-Assisted Classification","authors":"Cao Yu-kun, Wei Zi-yue, Tang Yi-jia, Jin Cheng-kun","doi":"10.1109/DDCLS58216.2023.10166293","DOIUrl":"https://doi.org/10.1109/DDCLS58216.2023.10166293","url":null,"abstract":"Hierarchical label text classification is a challenging task in the field of natural language processing, where each document needs to be correctly classified into multiple labels with hierarchical structure. However, in the label set, due to the insufficient semantic information contained in the labels and the small number of documents classified under deep-level labels, the training of deep-level labels is insufficient, leading to a significant imbalance in label training. To address this, a hierarchical label text classification method with deep-level label-assisted classification (DLAC) is proposed. The method proposes a deep-level label-assisted classifier, which effectively utilizes text features and rich features of shallow label nodes corresponding to deep label nodes (i.e., shallow label's rich features) on the basis of enhanced label semantics to enhance the classification performance of deep labels. The comparison experiment results with eleven algorithms on three datasets show that the model can effectively improve the classification performance of deep-level labels and achieve good results.","PeriodicalId":415532,"journal":{"name":"2023 IEEE 12th Data Driven Control and Learning Systems Conference (DDCLS)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133091871","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：