2019 IEEE International Conferences on Ubiquitous Computing & Communications (IUCC) and Data Science and Computational Intelligence (DSCI) and Smart Computing, Networking and Services (SmartCNS)最新文献
Pub Date : 2019-10-01DOI: 10.1109/iucc/dsci/smartcns.2019.00019
Xuefeng Chen
The MLSys 2019 workshop promotes the in-depth exploration of the most recent theory and applications, as well as future machine learning research of/for smart system fields. This workshop provides a forum to discuss the series of machine learning relevant issues, including machine learning based system model, machine learning based planning and designing on railway system, machine learning assisted analysis and construction on railway system, machine learning based analysis and application on IoT system, machine learning based control and energy management of electrified vehicles, machine learning based vehicular sensors and intelligent transportation systems, deep reinforcement learning enabled radio resource management in satellite communications, machine learning based system model, transfer learning enabled wireless communications, smart spectrum sensing and sharing powered by machine learning, machine learning based analysis and application on smart grid, machine learning for LTE/5G mobile system, etc.
{"title":"Message from the MLSys 2019 Workshop Chairs","authors":"Xuefeng Chen","doi":"10.1109/iucc/dsci/smartcns.2019.00019","DOIUrl":"https://doi.org/10.1109/iucc/dsci/smartcns.2019.00019","url":null,"abstract":"The MLSys 2019 workshop promotes the in-depth exploration of the most recent theory and applications, as well as future machine learning research of/for smart system fields. This workshop provides a forum to discuss the series of machine learning relevant issues, including machine learning based system model, machine learning based planning and designing on railway system, machine learning assisted analysis and construction on railway system, machine learning based analysis and application on IoT system, machine learning based control and energy management of electrified vehicles, machine learning based vehicular sensors and intelligent transportation systems, deep reinforcement learning enabled radio resource management in satellite communications, machine learning based system model, transfer learning enabled wireless communications, smart spectrum sensing and sharing powered by machine learning, machine learning based analysis and application on smart grid, machine learning for LTE/5G mobile system, etc.","PeriodicalId":410905,"journal":{"name":"2019 IEEE International Conferences on Ubiquitous Computing & Communications (IUCC) and Data Science and Computational Intelligence (DSCI) and Smart Computing, Networking and Services (SmartCNS)","volume":"312 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131836180","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 : 2019-10-01DOI: 10.1109/IUCC/DSCI/SmartCNS.2019.00146
Yongjie Gui, Hedan Liu, Hang Yin, Zhongzhi Li, Na Guo
In order to improve the development efficiency of Hex game computer game program, the program which can load different game engines and carry out human-machine competitiveas a test platform in the development process is designed and realized. Through pipeline communication to realize the interaction between the platform and the game program, design the Hex game referee terminal logic. This paper introduces the function of the program, designs the goal, and lists the treatment of the key problems in the program. The practical application proves that the game program runs fast and stable, and the design and realization of the man-machine game program has certain reference value to complete other game computer game programs.
{"title":"Design and Implementation of Hex Computer Game Platform","authors":"Yongjie Gui, Hedan Liu, Hang Yin, Zhongzhi Li, Na Guo","doi":"10.1109/IUCC/DSCI/SmartCNS.2019.00146","DOIUrl":"https://doi.org/10.1109/IUCC/DSCI/SmartCNS.2019.00146","url":null,"abstract":"In order to improve the development efficiency of Hex game computer game program, the program which can load different game engines and carry out human-machine competitiveas a test platform in the development process is designed and realized. Through pipeline communication to realize the interaction between the platform and the game program, design the Hex game referee terminal logic. This paper introduces the function of the program, designs the goal, and lists the treatment of the key problems in the program. The practical application proves that the game program runs fast and stable, and the design and realization of the man-machine game program has certain reference value to complete other game computer game programs.","PeriodicalId":410905,"journal":{"name":"2019 IEEE International Conferences on Ubiquitous Computing & Communications (IUCC) and Data Science and Computational Intelligence (DSCI) and Smart Computing, Networking and Services (SmartCNS)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125602163","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}
The living environment monitoring is a critical issue today. Many related researches focused on developing the living environment monitoring systems by using the web-based technology, but such information systems would increase the difficulty of promotion and migration to others, because of needing well-trained engineers to achieve and maintain such information system. This research focuses on developing the weather monitoring system. The main goal of the research is to use the technology of message queuing telemetry transport (MQTT) as roles of the communication layer instead of direct-connecting database, which can isolate the system migration complexity from heterogeneous relational database management system (RDBMS) and construct a distributed information system easily. Finally the contributions are demonstrated as implementation of the IoT-based weather monitoring system. The prototyping of the weather monitoring system is implemented and deployed at Wu-Tso elementary school for the students' learning of nature science and activity environment. In addition, the system is implemented using open sources and is easy to be deployed and scalable.
{"title":"An Implementation of IoT-Based Weather Monitoring System","authors":"Yu-Chung Tsao, Yin-Te Tsai, Yaw-Wen Kuo, Chaokung Hwang","doi":"10.1109/IUCC/DSCI/SmartCNS.2019.00135","DOIUrl":"https://doi.org/10.1109/IUCC/DSCI/SmartCNS.2019.00135","url":null,"abstract":"The living environment monitoring is a critical issue today. Many related researches focused on developing the living environment monitoring systems by using the web-based technology, but such information systems would increase the difficulty of promotion and migration to others, because of needing well-trained engineers to achieve and maintain such information system. This research focuses on developing the weather monitoring system. The main goal of the research is to use the technology of message queuing telemetry transport (MQTT) as roles of the communication layer instead of direct-connecting database, which can isolate the system migration complexity from heterogeneous relational database management system (RDBMS) and construct a distributed information system easily. Finally the contributions are demonstrated as implementation of the IoT-based weather monitoring system. The prototyping of the weather monitoring system is implemented and deployed at Wu-Tso elementary school for the students' learning of nature science and activity environment. In addition, the system is implemented using open sources and is easy to be deployed and scalable.","PeriodicalId":410905,"journal":{"name":"2019 IEEE International Conferences on Ubiquitous Computing & Communications (IUCC) and Data Science and Computational Intelligence (DSCI) and Smart Computing, Networking and Services (SmartCNS)","volume":"114 3-4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131450539","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 : 2019-10-01DOI: 10.1109/iucc/dsci/smartcns.2019.00008
{"title":"Message from the DSCI 2019 Program Chairs","authors":"","doi":"10.1109/iucc/dsci/smartcns.2019.00008","DOIUrl":"https://doi.org/10.1109/iucc/dsci/smartcns.2019.00008","url":null,"abstract":"","PeriodicalId":410905,"journal":{"name":"2019 IEEE International Conferences on Ubiquitous Computing & Communications (IUCC) and Data Science and Computational Intelligence (DSCI) and Smart Computing, Networking and Services (SmartCNS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134086723","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}
The social groups often represent the groups within which are dense connections and between which are sparse connections. The social groups are actually the clusters. In this paper, we propose a simple but powerful method to combine the content and link information of the social network in the class, and analyze the results of different graph clustering algorithms, our experimental result shows that social groups detection by Affinity Propagation algorithm outperforms than the other clustering algorithms. In addition, we analyze the centrality of the detected communities and find the groups detected can significantly help to improve the quality of the teaching and learning.
{"title":"Detection of Social Groups in Class by Affinity Propagation","authors":"Yajie Wang, Qiyang Peng, Zhao Pei, Miao Ma, Yuli Chen, Chengcai Leng, Honghong Yang","doi":"10.1109/IUCC/DSCI/SmartCNS.2019.00106","DOIUrl":"https://doi.org/10.1109/IUCC/DSCI/SmartCNS.2019.00106","url":null,"abstract":"The social groups often represent the groups within which are dense connections and between which are sparse connections. The social groups are actually the clusters. In this paper, we propose a simple but powerful method to combine the content and link information of the social network in the class, and analyze the results of different graph clustering algorithms, our experimental result shows that social groups detection by Affinity Propagation algorithm outperforms than the other clustering algorithms. In addition, we analyze the centrality of the detected communities and find the groups detected can significantly help to improve the quality of the teaching and learning.","PeriodicalId":410905,"journal":{"name":"2019 IEEE International Conferences on Ubiquitous Computing & Communications (IUCC) and Data Science and Computational Intelligence (DSCI) and Smart Computing, Networking and Services (SmartCNS)","volume":"68 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131038291","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 : 2019-10-01DOI: 10.1109/IUCC/DSCI/SmartCNS.2019.00032
Zhenzhou Guo, Shan Wang, Weifeng Jin, Changqing Gong, Na Lin
Protection of privacy has become an essential problem in cloud platform security. In many cases, data is shared with the third party for the analysis purpose. However, the sharing of data for analysis is not safe. Fully homomorphic encryption (FHE) is very promising to deal with ciphertext without decryption, FHE has become one of the key technologies to improve the security of user sensitive information. In this paper, we solve the problem of privacy preserving k-Nearest neighbor classification (K-NN), which forms the basis of many data analysis applications. We propose a scheme FK-NN, which is based on homomorphic encryption and numerical comparator. In our scheme, the homomorphic subtraction operation is designed and implemented firstly. Then, the cloud calculates the nearest neighbors of a given data point while the data point as well as the data points in the training set are in encrypted form. We can obtain classification results which are in encrypted form. The correctness of the scheme has been shown over cardiac disease dataset. The results show the efficiency of the proposed scheme satisfy the requirements of the system and accurately classify data which is in encrypted form.
{"title":"A k-Nearest Neighbor Algorithm Based on Homomorphic Encryption","authors":"Zhenzhou Guo, Shan Wang, Weifeng Jin, Changqing Gong, Na Lin","doi":"10.1109/IUCC/DSCI/SmartCNS.2019.00032","DOIUrl":"https://doi.org/10.1109/IUCC/DSCI/SmartCNS.2019.00032","url":null,"abstract":"Protection of privacy has become an essential problem in cloud platform security. In many cases, data is shared with the third party for the analysis purpose. However, the sharing of data for analysis is not safe. Fully homomorphic encryption (FHE) is very promising to deal with ciphertext without decryption, FHE has become one of the key technologies to improve the security of user sensitive information. In this paper, we solve the problem of privacy preserving k-Nearest neighbor classification (K-NN), which forms the basis of many data analysis applications. We propose a scheme FK-NN, which is based on homomorphic encryption and numerical comparator. In our scheme, the homomorphic subtraction operation is designed and implemented firstly. Then, the cloud calculates the nearest neighbors of a given data point while the data point as well as the data points in the training set are in encrypted form. We can obtain classification results which are in encrypted form. The correctness of the scheme has been shown over cardiac disease dataset. The results show the efficiency of the proposed scheme satisfy the requirements of the system and accurately classify data which is in encrypted form.","PeriodicalId":410905,"journal":{"name":"2019 IEEE International Conferences on Ubiquitous Computing & Communications (IUCC) and Data Science and Computational Intelligence (DSCI) and Smart Computing, Networking and Services (SmartCNS)","volume":"103 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115751278","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 : 2019-10-01DOI: 10.1109/IUCC/DSCI/SmartCNS.2019.00112
Xuancheng Guo, Hui Lin, Chuanfeng Xu, Wenzhong Lin
Recent advances in the healthcare system of Internet of Things (IoT) has led to a generation of a large amount of physic sensor data. Data analyst collects and analyzes these sensor data through wireless sensor network, so as to provide some treatment advices to physicians and patients. As a common data mining method, the k-means clustering algorithm is being applied to process large-scale sensor data. However, it also poses a threat of privacy leakage in the specific application process. To enhance the privacy in healthcare system of IoT, mutual privacypreserving k-means strategy (M-PPKS) based on homomorphic encryption is proposed in this paper, which neither discloses an individual's private information nor leaks the cluster center's characteristic data. An extension performance evaluation shows that, in the case of ensuring accurate clustering results, even if the analyst and individuals collude, the M-PPKS can prevent the disclosure of private information.
{"title":"A Data Clustering Strategy for Enhancing Mutual Privacy in Healthcare System of IoT","authors":"Xuancheng Guo, Hui Lin, Chuanfeng Xu, Wenzhong Lin","doi":"10.1109/IUCC/DSCI/SmartCNS.2019.00112","DOIUrl":"https://doi.org/10.1109/IUCC/DSCI/SmartCNS.2019.00112","url":null,"abstract":"Recent advances in the healthcare system of Internet of Things (IoT) has led to a generation of a large amount of physic sensor data. Data analyst collects and analyzes these sensor data through wireless sensor network, so as to provide some treatment advices to physicians and patients. As a common data mining method, the k-means clustering algorithm is being applied to process large-scale sensor data. However, it also poses a threat of privacy leakage in the specific application process. To enhance the privacy in healthcare system of IoT, mutual privacypreserving k-means strategy (M-PPKS) based on homomorphic encryption is proposed in this paper, which neither discloses an individual's private information nor leaks the cluster center's characteristic data. An extension performance evaluation shows that, in the case of ensuring accurate clustering results, even if the analyst and individuals collude, the M-PPKS can prevent the disclosure of private information.","PeriodicalId":410905,"journal":{"name":"2019 IEEE International Conferences on Ubiquitous Computing & Communications (IUCC) and Data Science and Computational Intelligence (DSCI) and Smart Computing, Networking and Services (SmartCNS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128692226","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 : 2019-10-01DOI: 10.1109/IUCC/DSCI/SmartCNS.2019.00160
Hua Li, Xiangxue Kong, Wen Liu
Cloud computing permits a significant improvement of the computational performance due to its integration of resources. Virtualization technology has an important impact on how the resources are integrated. In cloud computing, virtualization of Graphics Processing Unit (GPU) is the key technology for GPU intensive workloads such as game rendering, film production, etc. This paper proposes a strategy of the global illumination, specifically the effect of caustic of transparent geometries in virtualized framework platform. An GPU scheduling method (DGC) on different VMs was proposed to reduce the computation complexity. The feasibility of the algorithm is verified by ray tracing algorithm under OptiX frame work. We demonstrate the effectiveness of our approach with several simulated and fabricated examples under the virtualization framework.
云计算由于其对资源的集成，使得计算性能得到了显著的提高。虚拟化技术对如何集成资源有重要的影响。在云计算中，图形处理单元(Graphics Processing Unit, GPU)虚拟化是解决游戏渲染、电影制作等GPU密集型工作负载的关键技术。本文提出了一种全局照明策略，特别是在虚拟框架平台中透明几何图形的焦散效应。为了降低计算复杂度，提出了一种基于不同虚拟机的GPU调度方法(DGC)。通过OptiX框架下的光线追踪算法验证了该算法的可行性。在虚拟化框架下，我们用几个模拟和虚构的例子证明了我们方法的有效性。
{"title":"Dynamic Caustic Generation for Transparent Geometry in Virtualized GPU Environment","authors":"Hua Li, Xiangxue Kong, Wen Liu","doi":"10.1109/IUCC/DSCI/SmartCNS.2019.00160","DOIUrl":"https://doi.org/10.1109/IUCC/DSCI/SmartCNS.2019.00160","url":null,"abstract":"Cloud computing permits a significant improvement of the computational performance due to its integration of resources. Virtualization technology has an important impact on how the resources are integrated. In cloud computing, virtualization of Graphics Processing Unit (GPU) is the key technology for GPU intensive workloads such as game rendering, film production, etc. This paper proposes a strategy of the global illumination, specifically the effect of caustic of transparent geometries in virtualized framework platform. An GPU scheduling method (DGC) on different VMs was proposed to reduce the computation complexity. The feasibility of the algorithm is verified by ray tracing algorithm under OptiX frame work. We demonstrate the effectiveness of our approach with several simulated and fabricated examples under the virtualization framework.","PeriodicalId":410905,"journal":{"name":"2019 IEEE International Conferences on Ubiquitous Computing & Communications (IUCC) and Data Science and Computational Intelligence (DSCI) and Smart Computing, Networking and Services (SmartCNS)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117088666","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 : 2019-10-01DOI: 10.1109/IUCC/DSCI/SmartCNS.2019.00069
Lei Qiu, Yuan Li, Jing Sun, Jinsheng Liu, Yuhai Zhao
Genetic association study (GAS) is crucial to reveal the underlying principles of complex diseases. This is the first work that introduces contrast subgraph mining in two networks with significantly different edges or edge weights but the same vertices to solve GAS problem. It can not only detect the genetic loci highly associated with certain diseases, but discriminate between disease causing and disease preventing genetic loci, which captures more comprehensive and informative value for biologists. Inspired by the concept of r-core and minimum vertex weight, we propose to identify the novel global contrast r-core subgraphs r-GCCSs, which is more robust to outliers and redundancy. Further, we formulate the top-k r-GCCSs detection problem based on global contrast measure. In particular, (1) a linear time search algorithm is carefully developed to find the top-k r-GCCSs; (2) To further reduce the high computational cost, a linear space index is devised to support the top-k search. Comprehensive experiments on four large-scale real datasets demonstrate the efficiency and effectiveness of our approaches.
{"title":"Finding Global Contrast Core Subgraphs in Large-Scale Genetic Association Study","authors":"Lei Qiu, Yuan Li, Jing Sun, Jinsheng Liu, Yuhai Zhao","doi":"10.1109/IUCC/DSCI/SmartCNS.2019.00069","DOIUrl":"https://doi.org/10.1109/IUCC/DSCI/SmartCNS.2019.00069","url":null,"abstract":"Genetic association study (GAS) is crucial to reveal the underlying principles of complex diseases. This is the first work that introduces contrast subgraph mining in two networks with significantly different edges or edge weights but the same vertices to solve GAS problem. It can not only detect the genetic loci highly associated with certain diseases, but discriminate between disease causing and disease preventing genetic loci, which captures more comprehensive and informative value for biologists. Inspired by the concept of r-core and minimum vertex weight, we propose to identify the novel global contrast r-core subgraphs r-GCCSs, which is more robust to outliers and redundancy. Further, we formulate the top-k r-GCCSs detection problem based on global contrast measure. In particular, (1) a linear time search algorithm is carefully developed to find the top-k r-GCCSs; (2) To further reduce the high computational cost, a linear space index is devised to support the top-k search. Comprehensive experiments on four large-scale real datasets demonstrate the efficiency and effectiveness of our approaches.","PeriodicalId":410905,"journal":{"name":"2019 IEEE International Conferences on Ubiquitous Computing & Communications (IUCC) and Data Science and Computational Intelligence (DSCI) and Smart Computing, Networking and Services (SmartCNS)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122103637","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}
Text Classification is a hot topic in natural language processing. In view of the strong correlation the structure of natural language, direct translation the text into vector will lead to too high dimension. In addition, traditional word vector usually maps words with a single vector, which cannot represent the polyseme of words and affect the accuracy of the final classification. In this paper, we propose a novel DCNN-BiGRU (Deep Convolutional Neural Network Bidirection Gated Recurrent) text classification model based on BERT(Bidirectional Encoder Representations from Transformer) embedding. The model adopts the BERT to train the language model of word semantic representation. The semantic vector is generated dynamically according to the context of the word, and then it is put into the DCNN-BiGRU hybrid model. By doing so, the semantic vector not only contains the local features of text but also the context features of text. Experiments on CCERT Chinese email sample set and movie comment data set verify the validity of this model.
{"title":"DCNN-BiGRU Text Classification Model Based on BERT Embedding","authors":"He Huang, Xiaoyuan Jing, Fei Wu, Yong-Fang Yao, Xinyu Zhang, Xiwei Dong","doi":"10.1109/IUCC/DSCI/SmartCNS.2019.00132","DOIUrl":"https://doi.org/10.1109/IUCC/DSCI/SmartCNS.2019.00132","url":null,"abstract":"Text Classification is a hot topic in natural language processing. In view of the strong correlation the structure of natural language, direct translation the text into vector will lead to too high dimension. In addition, traditional word vector usually maps words with a single vector, which cannot represent the polyseme of words and affect the accuracy of the final classification. In this paper, we propose a novel DCNN-BiGRU (Deep Convolutional Neural Network Bidirection Gated Recurrent) text classification model based on BERT(Bidirectional Encoder Representations from Transformer) embedding. The model adopts the BERT to train the language model of word semantic representation. The semantic vector is generated dynamically according to the context of the word, and then it is put into the DCNN-BiGRU hybrid model. By doing so, the semantic vector not only contains the local features of text but also the context features of text. Experiments on CCERT Chinese email sample set and movie comment data set verify the validity of this model.","PeriodicalId":410905,"journal":{"name":"2019 IEEE International Conferences on Ubiquitous Computing & Communications (IUCC) and Data Science and Computational Intelligence (DSCI) and Smart Computing, Networking and Services (SmartCNS)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129148705","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}
2019 IEEE International Conferences on Ubiquitous Computing & Communications (IUCC) and Data Science and Computational Intelligence (DSCI) and Smart Computing, Networking and Services (SmartCNS)
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：