检索结果-内蒙古大学图书馆

IEEE International Conference on Big Data and Cloud computing (BdCloud)

作者： Hongzhong Wang Kun Guo Zhanghui Liu Fujian Provincial Key Laboratory of Network Computing and Intelligent Information Processing College of Mathematics and Computer Science Fuzhou University Fuzhou China Key Laboratory of Spatial Data Mining & Information Sharing Ministry of Education Fuzhou China College of Mathematics and Computer Science Fuzhou University Fuzhou China

ISBN: (数字)9781728143286

ISBN: (纸本)9781728143293

This paper presents a self-training word embedding text classification model based on knowledge graph expansion for text classification. Current mixed word embedding methods are overly dependent on the Fasttext pre-training model and here is still a problem of missing words with rich semantic information are not mapped. First, we propose a method for extracting missing nouns based on shape near word filtering. Second, we design a self-training word embedding method based on knowledge graph that mixes with pre-training word embedding to obtain a high-quality mixed word vector with rich semantics and rich semantics. Third, we designed a GRU model based on improved mixed word embedding to improve the quality of text classification. Experiments conducted on multiple text classification datasets demonstrate that our methods can effectively improve the text classification accuracy.

关键词： Text categorization Context modeling Feature extraction Training Task analysis Predictive models

来源：评论

学校读者我要写书评

暂无评论

Financial Big Data Hot and Cold Separation Scheme Based on HBase and Redis

Financial Big Data Hot and Cold Separation Scheme Based on H...

引用

IEEE International Conference on Big Data and Cloud computing (BdCloud)

作者： Kunhui Li Kun Guo Hong Guo Fujian Provincial Key Laboratory of Network Computing and Intelligent Information Processing College of Mathematics and Computer Science Fuzhou University Fuzhou China Key Laboratory of Spatial Data Mining Information Sharing Ministry of Education Fuzhou China College of Mathematics and Computer Science Fuzhou University Fuzhou China

ISBN: (数字)9781728143286

ISBN: (纸本)9781728143293

In the era of big data, HBase has been widely used in scenarios of massive unstructured data. For the financial big data, due to the integrity and timing of it, unreasonable data storage and management usually lead to hot spots that decreases the query performance. In practice, the separation of hot and cold financial data will improve data query performance and utilization rate of cluster resources. In this paper, a hot and cold data separation scheme is designed, to store infrequently queried financial data to HBase, and frequently queried one to Redis. The cold data is reasonably planned and managed through pre-partitioning and row key design for HBase. A hot data cache based on Redis is realized to improve the query speed and reduces the pressure of HBase. In addition, due to the lack of Redis's inherent cache elimination strategy, we propose a caching strategy based on the frequencies of updating and querying operations. The experimental results show that the scheme can effectively avoid the hot storage problem, and improve the query performance, and improve the cache hit ratio of Redis. Therefore, the number of cold data access requests can be effectively reduced.

关键词： Big Data Distributed databases Memory Time series analysis Stock markets

来源：评论

学校读者我要写书评

暂无评论

Overlapping Community Detection Based on Influence and Seeds Extension

引用

Tien Tzu Hsueh Pao/Acta Electronica Sinica 2019年第1期47卷 153-160页

作者： Yu, Zhi-Yong Chen, Ji-Jie Guo, Kun Chen, Yu-Zhong Xu, Qian College of Mathematics and Computer Sciences Fuzhou University FuzhouFujian350116 China Fujian Provincial Key Laboratory of Network Computing and Intelligent Information Processing FuzhouFujian350116 China Ministry of Education Key Laboratory of Spatial Data Mining & Information Sharing FuzhouFujian350116 China State Grid Info-Telecom Great Power Science and Technology Co. Ltd. FuzhouFujian350003 China

Community detection is a significant research direction in the research of social networks. To improve the quality of seeds selection and expansion, we propose an influence seeds extension overlapping community detection (i-SEOCD) algorithm for overlapping community detection. First, i-SEOCD uses a node influence strategy to find the seed communities with tight structures. Second, on the basis of the seed communities, we calculate the similarity among communities and their neighbor nodes. The nodes whose similarity is greater than a predefined threshold are selected. Third, the strategy of optimizing a self-adaptive function is adopted to expand the communities. Finally, the free nodes in the network are assigned to their corresponding communities in order to find out all the overlapping community structures. Experiments on the real and artificial networks show that i-SEOCD is capable of discovering overlapping communities in complex social networks efficiently. © 2019, Chinese Institute of Electronics. All right reserved.

关键词： Population dynamics

来源：评论

学校读者我要写书评

暂无评论

A Dynamic Community Detection Algorithm Based on Parallel Incremental Related Vertices 2

A Dynamic Community Detection Algorithm Based on Parallel In...

引用

2017 IEEE 2nd International Conference on Big Data Analysis（ICBDA 2017）

作者： Guohui Li Kun Guo YuZhong Chen Ling Wu Danhong Zhu College of Mathematics and Computer Sciences Fuzhou University Fujian Provincial Key Laboratory of Network Computing and Intelligent Information Processing(Fuzhou University) Key Laboratory of Spatial Data Mining & Information Sharing Ministry of Education

ISBN: (纸本)9781509036202

One of the traditional ways for detecting dynamic communities is to find the communities at each interval through the static community detection ***,it usually leads to high computation *** this paper,a novel algorithm based on the Map Reduce model and the label propagation progress with the strategy of incremental related vertices is proposed,which is called PLPIRV(Parallel Label Propagation and Incremental Related Vertices).Based on the communities found at the previous interval,the new algorithm adjusts the communities the incremental related vertices belong *** clustering of the whole network can be avoided by incrementally analyzing the variation of the networks,so that the time cost can be greatly *** on artificial and real datasets show that the proposed algorithm performs well on dynamic community detection.

关键词： dynamic community detection parallel computation incremental algorithm label propagation

来源：评论

学校读者我要写书评

暂无评论

Parallel Multi-Label Propagation Based on Influence Model and Its Application to Overlapping Community Discovery

Parallel Multi-Label Propagation Based on Influence Model an...

引用

作者： Qiu, Qirong Guo, Wenzhong Chen, Yuzhong Guo, Kun Li, Rongrong School of Economics and Management Fuzhou University Fujian Provincial Key Laboratory of Network Computing and Intelligent Information Processing Fuzhou350116 China College of Mathematics and Computer Science Fuzhou University Fujian Provincial Key Laboratory of Network Computing and Intelligent Information Processing Key Laboratory of Spatial Data Mining and Information Sharing Ministry of Education Fuzhou350116 China

Finding communities in networks is one of the challenging issues in complex network research. We have to deal with very large networks that contain billions of vertices, which makes community discovery a computationally intensive work. Moreover, communities usually overlap each other, which greatly increases the difficulty of identifying the boundaries of communities. In this paper, we propose a parallel multi-label propagation algorithm (PMLPA) that enhances traditional multi-label propagation algorithm (MLPA) in two ways. First, the critical steps of MLPA are parallelized based on the MapReduce model to get higher scalability. Second, new label updating strategy is used to automatically determine the most valuable labels of each vertex. Furthermore, we study the improvement of PMLPA through considering the influence of vertices and labels on label updating. In this way, the importance of each label can be described with higher precision. Experiments on artificial and real networks prove that the proposed algorithms can achieve both high discovering accuracy and high scalability. © 2017 World Scientific Publishing Company.

关键词： Complex networks

来源：评论

学校读者我要写书评

暂无评论

Self-Evolutionary Neuron Model for Fast-Response Spiking Neural networks

TechRxiv

引用

TechRxiv 2019年

作者： Zhang, Anguo Han, Ying Hu, Jing Niu, Yuzhen Gao, Yueming Chen, Zhizhang Zhao, Kai College of Physics and Information Engineering Fuzhou University Fujian350108 China The Key Laboratory of Medical Instrumentation Pharmaceutical Technology of Fujian Province Fuzhou350116 China Research Institute of Ruijie Ruijie Networks Co. Ltd Fujian350002 China School of Public Health Xiamen University Xiamen361102 China College of Information and Intelligent Transportation Fujian Chuanzheng Communications College Fuzhou350007 China Fujian Key Laboratory of Network Computing and Intelligent Information Processing College of Mathematics and Computer Science Fuzhou University The Key Laboratory of Spatial Data Mining and Information Sharing Ministry of Education Fujian350108 China College of Physics and information Engineering Fuzhou University the Key Laboratory of Medical Instrumentation Pharmaceutical Technology of Fujian Province Fujian350108 China College of Physics and Information Engineering Fuzhou University 350108 China Faculty of Science and Technology University of Macau 999078 China

We propose two simple and effective spiking neuron models to improve the response time of the conventional spiking neural network. The proposed neuron models adaptively tune the presynaptic input current depending on the input received from its presynapses and subsequent neuron firing events. We analyze and derive the firing activity homeostatic convergence of the proposed models. We experimentally verify and compare the models on MNIST handwritten digits and FashionMNIST classification tasks. We show that the proposed neuron models significantly increase the response speed to the input signal. Experiment codes are available at https://***/anvien/Evol-SNN. © 2019, CC0.

关键词： Neural networks

来源：评论

学校读者我要写书评

暂无评论

Encoding-Aware Data Placement for Efficient Degraded Reads in XOR-Coded Storage Systems

Encoding-Aware Data Placement for Efficient Degraded Reads i...

引用

Reliable Distributed Systems

作者： Zhirong Shen Patrick P.C. Lee Jiwu Shu Wenzhong Guo College of Mathematics and Computer Science Key Laboratory of Network Computing and Intelligent Information Processing Fuzhou China Department of Computer Science and Engineering The Chinese University of Hong Kong Department of Computer Science and Technology Tsinghua University Key Laboratory of Spatial Data Mining and Information Sharing Minstry of Education Fuzhou China

ISBN: (纸本)9781509035144

Erasure coding has been increasingly used by distributed storage systems to maintain fault tolerance with low storage redundancy. However, how to enhance the performance of degraded reads in erasure-coded storage has been a critical issue. We revisit this problem from two different perspectives that are neglected by existing studies: data placement and encoding rules. To this end, we propose an encoding-aware data placement (EDP) approach that aims to reduce the number of I/Os in degraded reads during a single failure for general XOR-based erasure codes. EDP carefully places sequential data based on the encoding rules of the given erasure code. Trace-driven evaluation results show that compared to two baseline data placement methods, EDP reduces up to 37.4% of read data on the most loaded disk and shortens up to 15.4% of read time.

关键词： Encoding Distributed databases Layout Fault tolerant systems Redundancy Computer science

来源：评论

学校读者我要写书评

暂无评论

A nonlocal interferometric phase filtering strategy for practical InSAR system

A nonlocal interferometric phase filtering strategy for prac...

引用

IEEE International Symposium on Geoscience and Remote Sensing (IGARSS)

作者： Xue Lin Dongsheng Fang Fangfang Li Ke Wang Shandong Provincial Key Laboratory of Network Based Intelligent Computing Jinan China School of Information Science and Engineering University of Jinan Jinan China Institute of Electronics Chinese Academy of Sciences Beijing China University of Chinese Academy of Sciences Beijing China Key Laboratory of Technology in Geo-spatial Information Processing and Application System Chinese Academy of Sciences Beijing China School of Information Science and Engineering Henan University of Technology Zhengzhou China

Nonlocal interferometric phase filtering methods achieve excellent performance in both noise reduction and texture preservation, even in the case of complicated topography and low coherence. The main limitation of the nonlocal methods is the computational burden. This paper proposed a nonlocal phase filtering strategy for the practical InSAR system, which combine the nonlocal algorithm with the traditional method to improve the efficiency.

关键词： Filtering Coherence Noise reduction Frequency estimation Estimation Algorithm design and analysis Surfaces

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：