Pub Date : 2023-03-01DOI: 10.1109/ICNLP58431.2023.00083
Wang Peng, Chunhui Hu
In the retrieval process of industrial models, the traditional database retrieval can no longer meet their needs in terms of efficiency and precision because of their multi-source heterogeneous, complex types and large information scale. This paper optimizes the Elasticsearch search engine in three aspects: the underlying index of the search engine, keyword search and sorting algorithm; and verifies the feasibility of the method through experiments.
{"title":"Research on Optimization Methods for Industrial Model Retrieval","authors":"Wang Peng, Chunhui Hu","doi":"10.1109/ICNLP58431.2023.00083","DOIUrl":"https://doi.org/10.1109/ICNLP58431.2023.00083","url":null,"abstract":"In the retrieval process of industrial models, the traditional database retrieval can no longer meet their needs in terms of efficiency and precision because of their multi-source heterogeneous, complex types and large information scale. This paper optimizes the Elasticsearch search engine in three aspects: the underlying index of the search engine, keyword search and sorting algorithm; and verifies the feasibility of the method through experiments.","PeriodicalId":53637,"journal":{"name":"Icon","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79758577","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 : 2023-03-01DOI: 10.1109/ICNLP58431.2023.00010
Mengyang Cheng, Haibo Ge, Sai Ma, Wenhao He, Yu An, Ting Zhou
In order to solve the problem of missed detection and false detection of small targets in object detection, and to improve the detection precision and recall of small object, this paper proposes a small object detection algorithm that introduces context information and attention mechanism. The algorithm is improved on the Faster RCNN network architecture, and a multilevel feature fusion module is proposed to solve the problem of incomplete extraction of detailed information. The proposed regional attention module solves the interference of background noise and focuses on the target to be detected. At the same time, in order to more effectively meet the characteristics of small target detection, we have improved the anchor box. The method proposed in this paper is verified on DIOR, PASCAL VOC2007 and MS COCO datasets. Experiments show that the algorithm proposed in this paper and the current advanced algorithm have better accuracy and precision in detecting small targets.
{"title":"Small Object Detection Based on Context Information and Attention Mechanism","authors":"Mengyang Cheng, Haibo Ge, Sai Ma, Wenhao He, Yu An, Ting Zhou","doi":"10.1109/ICNLP58431.2023.00010","DOIUrl":"https://doi.org/10.1109/ICNLP58431.2023.00010","url":null,"abstract":"In order to solve the problem of missed detection and false detection of small targets in object detection, and to improve the detection precision and recall of small object, this paper proposes a small object detection algorithm that introduces context information and attention mechanism. The algorithm is improved on the Faster RCNN network architecture, and a multilevel feature fusion module is proposed to solve the problem of incomplete extraction of detailed information. The proposed regional attention module solves the interference of background noise and focuses on the target to be detected. At the same time, in order to more effectively meet the characteristics of small target detection, we have improved the anchor box. The method proposed in this paper is verified on DIOR, PASCAL VOC2007 and MS COCO datasets. Experiments show that the algorithm proposed in this paper and the current advanced algorithm have better accuracy and precision in detecting small targets.","PeriodicalId":53637,"journal":{"name":"Icon","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81356683","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 : 2023-03-01DOI: 10.1109/icnlp58431.2023.00071
Xihai Xie, Biao Hui
To address the problem of capturing correlation peak amplitude decay and difficulty in determining the capture threshold when the demodulated signal is synchronously captured by DMF (all-digital matched filter) under the influence of Doppler frequency bias, the dichotomous DMF method is proposed: the pseudo-random sequence is segmented before and after to reduce the integration time of correlation operation, and then reduce the loss of normalized correlation peak under the influence of frequency bias integration decay. Matlab platform simulation verifies the effect of correlation length on the loss of correlation peaks under the frequency bias scenario, and the experimental results show that the dichotomous DMF method is less sensitive to the Doppler frequency bias than the DMF method. In order to reduce the hardware resource overhead of the pseudocode phase search module and to ensure a certain search efficiency, a serial-parallel structure with serial data transmission and parallel operation is adopted, which can search multiple code elements at a time for phase deviation in the pseudocode phase search and improve the capture efficiency.
{"title":"A fast capture structure for dichotomous DMF pseudocode based on DSP Builder","authors":"Xihai Xie, Biao Hui","doi":"10.1109/icnlp58431.2023.00071","DOIUrl":"https://doi.org/10.1109/icnlp58431.2023.00071","url":null,"abstract":"To address the problem of capturing correlation peak amplitude decay and difficulty in determining the capture threshold when the demodulated signal is synchronously captured by DMF (all-digital matched filter) under the influence of Doppler frequency bias, the dichotomous DMF method is proposed: the pseudo-random sequence is segmented before and after to reduce the integration time of correlation operation, and then reduce the loss of normalized correlation peak under the influence of frequency bias integration decay. Matlab platform simulation verifies the effect of correlation length on the loss of correlation peaks under the frequency bias scenario, and the experimental results show that the dichotomous DMF method is less sensitive to the Doppler frequency bias than the DMF method. In order to reduce the hardware resource overhead of the pseudocode phase search module and to ensure a certain search efficiency, a serial-parallel structure with serial data transmission and parallel operation is adopted, which can search multiple code elements at a time for phase deviation in the pseudocode phase search and improve the capture efficiency.","PeriodicalId":53637,"journal":{"name":"Icon","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85992728","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 : 2023-03-01DOI: 10.1109/ICNLP58431.2023.00084
Boyu Chen, Hongjie Liu, Lei Yin
The centralized data storage mode adopted by the traditional supply chain management system has some problems, such as single point of failure, data privacy disclosure, opaque internal operation of the system and so on, which seriously restricts the information flow and data sharing among enterprises. Blockchain is distributed, open, transparent and tamper proof. It can provide reliable underlying services for the implementation of distributed data security sharing system. Therefore, this paper proposes a blockchain based supply chain quality data security sharing model, which takes the distributed blockchain network as the core to build decentralized data security sharing services. At the same time, the practical Byzantine fault-tolerant (pbft) algorithm used in blockchain has some problems, such as large consensus delay, low throughput, poor performance, and does not support node dynamic management. Combined with the characteristics of supply chain alliance chain, by introducing simplified consistency protocol and new node management mechanism, the communication complexity of the algorithm is reduced and the dynamic management of nodes is realized.
{"title":"Blockchain for Supply Chain Data Security Sharing Consensus Algorithm Design","authors":"Boyu Chen, Hongjie Liu, Lei Yin","doi":"10.1109/ICNLP58431.2023.00084","DOIUrl":"https://doi.org/10.1109/ICNLP58431.2023.00084","url":null,"abstract":"The centralized data storage mode adopted by the traditional supply chain management system has some problems, such as single point of failure, data privacy disclosure, opaque internal operation of the system and so on, which seriously restricts the information flow and data sharing among enterprises. Blockchain is distributed, open, transparent and tamper proof. It can provide reliable underlying services for the implementation of distributed data security sharing system. Therefore, this paper proposes a blockchain based supply chain quality data security sharing model, which takes the distributed blockchain network as the core to build decentralized data security sharing services. At the same time, the practical Byzantine fault-tolerant (pbft) algorithm used in blockchain has some problems, such as large consensus delay, low throughput, poor performance, and does not support node dynamic management. Combined with the characteristics of supply chain alliance chain, by introducing simplified consistency protocol and new node management mechanism, the communication complexity of the algorithm is reduced and the dynamic management of nodes is realized.","PeriodicalId":53637,"journal":{"name":"Icon","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"91146587","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 : 2023-03-01DOI: 10.1109/ICNLP58431.2023.00075
Ping Liu, Xiangzhong Zeng, Shihao Gai, Hanning Sun
Wireless sensor network has been widely used as an important means of perceiving and monitoring the real environment. Node location algorithm is the key supporting technology for the normal operation of wireless sensor network nodes. To achieve higher positioning accuracy and improve the adaptability to the network, a beacon-based MSVR-DV-Hop (Multidimensional Support Vector Regression-DV-Hop) algorithm is proposed in three-dimensional scenes. The three stages of hop acquisition, distance estimation and coordinate calculation in classical DV-Hop algorithm are improved, and simulation experiments and result analysis are carried out in three-dimensional scene. The positioning accuracy of this algorithm is significantly improved compared with other algorithms in three-dimensional scenes, positioning error fluctuations are significantly improved in different anisotropic scenes, and positioning error fluctuations are stable in different anisotropic scenes, which has better adaptability and accuracy. Positioning errors in three-dimensional scenes are reduced by at least 56% compared to the classical three-dimensional DV-Hop algorithm and 12% compared to the LMSVR algorithm.
{"title":"Application Research of 3D MSVR-DV-Hop Algorithm Based on Node Filtering","authors":"Ping Liu, Xiangzhong Zeng, Shihao Gai, Hanning Sun","doi":"10.1109/ICNLP58431.2023.00075","DOIUrl":"https://doi.org/10.1109/ICNLP58431.2023.00075","url":null,"abstract":"Wireless sensor network has been widely used as an important means of perceiving and monitoring the real environment. Node location algorithm is the key supporting technology for the normal operation of wireless sensor network nodes. To achieve higher positioning accuracy and improve the adaptability to the network, a beacon-based MSVR-DV-Hop (Multidimensional Support Vector Regression-DV-Hop) algorithm is proposed in three-dimensional scenes. The three stages of hop acquisition, distance estimation and coordinate calculation in classical DV-Hop algorithm are improved, and simulation experiments and result analysis are carried out in three-dimensional scene. The positioning accuracy of this algorithm is significantly improved compared with other algorithms in three-dimensional scenes, positioning error fluctuations are significantly improved in different anisotropic scenes, and positioning error fluctuations are stable in different anisotropic scenes, which has better adaptability and accuracy. Positioning errors in three-dimensional scenes are reduced by at least 56% compared to the classical three-dimensional DV-Hop algorithm and 12% compared to the LMSVR algorithm.","PeriodicalId":53637,"journal":{"name":"Icon","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87459883","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 : 2023-03-01DOI: 10.1109/ICNLP58431.2023.00012
Zhijie Zhu, Jie Fang, Nan Wang, Jiaqiu Guan
In this paper, we propose a lightweight salient object detection framework called Multi-Constraint Coupling optimization Network (MCONet) to address the conflict between model scale and inference ability, which can learn more knowledge with fewer parameters through embedding feature priors. Specifically, we build a lightweight encoder as the backbone network to represent the image, and then use two parallel decoders to infer salient mask features and salient edge features respectively. Besides, we fuse the output features of different decoders by a convolutional block attention module (CBAM) module. In addition, we adopt a multi-constraint coupling optimization strategy to increase the soft constraints in the training phase, and improve the prior guidance of the edge to the inference results. Experimental results on 5 public benchmark datasets show that the proposed MCONet can reach comparable even better performance of state-of-the-art lightweight salient object detection models.
{"title":"Multi-constraint Coupling Optimization for Salient Object Detection","authors":"Zhijie Zhu, Jie Fang, Nan Wang, Jiaqiu Guan","doi":"10.1109/ICNLP58431.2023.00012","DOIUrl":"https://doi.org/10.1109/ICNLP58431.2023.00012","url":null,"abstract":"In this paper, we propose a lightweight salient object detection framework called Multi-Constraint Coupling optimization Network (MCONet) to address the conflict between model scale and inference ability, which can learn more knowledge with fewer parameters through embedding feature priors. Specifically, we build a lightweight encoder as the backbone network to represent the image, and then use two parallel decoders to infer salient mask features and salient edge features respectively. Besides, we fuse the output features of different decoders by a convolutional block attention module (CBAM) module. In addition, we adopt a multi-constraint coupling optimization strategy to increase the soft constraints in the training phase, and improve the prior guidance of the edge to the inference results. Experimental results on 5 public benchmark datasets show that the proposed MCONet can reach comparable even better performance of state-of-the-art lightweight salient object detection models.","PeriodicalId":53637,"journal":{"name":"Icon","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87602711","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 : 2023-03-01DOI: 10.1109/ICNLP58431.2023.00047
A. Ito, Kotaro Takeda, Shuichi Ishida
In this paper, we analyzed entrepreneurship-related text using the automatic personality trait estimation model to investigate the difference between entrepreneurship-related and other texts. We collected texts from fourteen participants of entre-preneurship-related classes and texts from different domains (impressions of other courses and tweets from Twitter). Next, we developed a personality estimation model using BERT and a feedforward network using Kaggle’s MBTI corpus. As a result of the analysis, we found significant personality differences between the entrepreneur-related and other texts in the judgment-perception dimension.
{"title":"Personality Analysis of Entrepreneurial Text for Entrepreneurship Education","authors":"A. Ito, Kotaro Takeda, Shuichi Ishida","doi":"10.1109/ICNLP58431.2023.00047","DOIUrl":"https://doi.org/10.1109/ICNLP58431.2023.00047","url":null,"abstract":"In this paper, we analyzed entrepreneurship-related text using the automatic personality trait estimation model to investigate the difference between entrepreneurship-related and other texts. We collected texts from fourteen participants of entre-preneurship-related classes and texts from different domains (impressions of other courses and tweets from Twitter). Next, we developed a personality estimation model using BERT and a feedforward network using Kaggle’s MBTI corpus. As a result of the analysis, we found significant personality differences between the entrepreneur-related and other texts in the judgment-perception dimension.","PeriodicalId":53637,"journal":{"name":"Icon","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74907102","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}
Camouflage Target Segmentation (COS) aims to segment targets hidden in complex environment. When the existing COS algorithm fuses multi-level features, it ignores the expression and positioning of the edge features of the camouflage target, and pays more attention to the influence of the fusion of features on the segmentation performance. Therefore, a COS algorithm based on disguised target segmentation based on reverse attention and self-interaction fusion is proposed. First, multi-scale features are extracted through the backbone network; Then, in order to improve the expression ability of edge features, the features extracted by the backbone network are enhanced using a network composed of a reverse attention module (RAM); Finally, the self-interaction fusion module (SIM) drives the features of different scales to achieve layer-by-layer fusion, while suppressing noise interference and obtaining more accurate target information. Experimental results show that on the three commonly used natural camouflage datasets of CHAMELEON, CAMO and CODIOK, the model shows better segmentation effect than other typical models.
{"title":"Camouflage target segmentation based on reverse attention and self-interaction fusion","authors":"Haibo Ge, Wenhao He, Yu An, Haodong Feng, Jiajun Geng, Chaofeng Huang","doi":"10.1109/ICNLP58431.2023.00015","DOIUrl":"https://doi.org/10.1109/ICNLP58431.2023.00015","url":null,"abstract":"Camouflage Target Segmentation (COS) aims to segment targets hidden in complex environment. When the existing COS algorithm fuses multi-level features, it ignores the expression and positioning of the edge features of the camouflage target, and pays more attention to the influence of the fusion of features on the segmentation performance. Therefore, a COS algorithm based on disguised target segmentation based on reverse attention and self-interaction fusion is proposed. First, multi-scale features are extracted through the backbone network; Then, in order to improve the expression ability of edge features, the features extracted by the backbone network are enhanced using a network composed of a reverse attention module (RAM); Finally, the self-interaction fusion module (SIM) drives the features of different scales to achieve layer-by-layer fusion, while suppressing noise interference and obtaining more accurate target information. Experimental results show that on the three commonly used natural camouflage datasets of CHAMELEON, CAMO and CODIOK, the model shows better segmentation effect than other typical models.","PeriodicalId":53637,"journal":{"name":"Icon","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75630534","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 : 2023-03-01DOI: 10.1109/ICNLP58431.2023.00068
Peng Xu, Guangyue Lu, Yuxin Li, Cai Xu
Intrusion detection in the Industrial Internet of Things (IIoT) is a challenge for the network security protection. Graph neural network (GNN) is employed to improve the network security by virtue of efficiently constructing a message passing function. However, existing intrusion detection methods based on GNN do not fully exploit the information of original data which results in the poor intrusion detection performance. In this paper, we propose an Exploiting Edge feature based on Graph Convolutional Network (EE-GCN), which can capture both the edge features of the network traffic link as well as the relationship between device nodes. In addition, we construct a two-layer GCN network to extract the edge features. Finally, two benchmark datasets (NF-BoT-IoT and NF-ToN-IoT) in Network Intrusion Detection System (NIDS) are used to evaluate the performance of the proposed method. The results show that the method proposed in this paper outperforms other methods.
{"title":"EE-GCN: A Graph Convolutional Network based Intrusion Detection Method for IIoT","authors":"Peng Xu, Guangyue Lu, Yuxin Li, Cai Xu","doi":"10.1109/ICNLP58431.2023.00068","DOIUrl":"https://doi.org/10.1109/ICNLP58431.2023.00068","url":null,"abstract":"Intrusion detection in the Industrial Internet of Things (IIoT) is a challenge for the network security protection. Graph neural network (GNN) is employed to improve the network security by virtue of efficiently constructing a message passing function. However, existing intrusion detection methods based on GNN do not fully exploit the information of original data which results in the poor intrusion detection performance. In this paper, we propose an Exploiting Edge feature based on Graph Convolutional Network (EE-GCN), which can capture both the edge features of the network traffic link as well as the relationship between device nodes. In addition, we construct a two-layer GCN network to extract the edge features. Finally, two benchmark datasets (NF-BoT-IoT and NF-ToN-IoT) in Network Intrusion Detection System (NIDS) are used to evaluate the performance of the proposed method. The results show that the method proposed in this paper outperforms other methods.","PeriodicalId":53637,"journal":{"name":"Icon","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81109430","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 the case that multiple service types of slices are jointly carried on core network, a Resource Management Algorithms Oriented to Slicing Functions (RMOSF) is proposed for the processing efficiency of slice requests and the resource allocation of the substrate network. First, the incoming slice requests are input into the admission control module, and the pre-accepted slice requests are screened out through the deep reinforcement learning algorithm; secondly, the pre-accepted slice requests are brought into the resource allocation module, and slices with different type are brought into the corresponding constrained optimization problems for solution; finally, when the substrate physical network resources are sufficient, the slices are mapped to start their life cycle. The simulation results show that the algorithm effectively improves slice profit and request acceptance rate, and also improves resource utilization.
{"title":"Resource Management Algorithm for Slicing Function in 5G Network Slicing","authors":"Jiawen Guo, Guohui Zhu, Dingyuan Zhang, Chenglin Xu","doi":"10.1109/ICNLP58431.2023.00073","DOIUrl":"https://doi.org/10.1109/ICNLP58431.2023.00073","url":null,"abstract":"In the case that multiple service types of slices are jointly carried on core network, a Resource Management Algorithms Oriented to Slicing Functions (RMOSF) is proposed for the processing efficiency of slice requests and the resource allocation of the substrate network. First, the incoming slice requests are input into the admission control module, and the pre-accepted slice requests are screened out through the deep reinforcement learning algorithm; secondly, the pre-accepted slice requests are brought into the resource allocation module, and slices with different type are brought into the corresponding constrained optimization problems for solution; finally, when the substrate physical network resources are sufficient, the slices are mapped to start their life cycle. The simulation results show that the algorithm effectively improves slice profit and request acceptance rate, and also improves resource utilization.","PeriodicalId":53637,"journal":{"name":"Icon","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79302251","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：