作者机构:
[张彬] Department of Computer Science, Hengyang Normal University, Hengyang 421008, Hunan, China;[蒋涛; Yue, Guangxue] College of Mathematics and Information Engineering, Jiaxing University, Jiaxing 314000, Zhejiang, China;[蒋涛; 李国徽] College of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan 430074, China
通讯机构:
Department of Computer Science, Hengyang Normal University, China
作者机构:
[蒋涛; 张彬] Department of Computer Science, Hengyang Normal University, Hengyang 421008, China;[朱虹; 蒋涛; 李国徽] College of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan 430074, China
通讯机构:
Department of Computer Science, Hengyang Normal University, China
作者机构:
[蒋涛; 张彬] Department of Computer Science, Hengyang Normal University, Hengyang 421008, China;[朱虹; 蒋涛; 李国徽] College of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan 430074, China
通讯机构:
Department of Computer Science, Hengyang Normal University, China
作者机构:
[旷海兰; 刘新华; 李方敏; 方艺霖] School of Information Engineering, Wuhan University of Technology, Wuhan 430070, China;[旷海兰] Computer Department of Hengyang Normal University, Hengyang, Hunan 421008, China
通讯机构:
School of Information Engineering, Wuhan University of Technology, China
作者机构:
[旷海兰; 李方敏; 方艺霖; 刘新华] School of Information Engineering, Wuhan University of Technology, Wuhan 430070, China;[旷海兰] Computer Department, Hengyang Normal University, Hengyang 421008, China
通讯机构:
School of Information Engineering, Wuhan University of Technology, China
关键词:
Data mining;Model;Data streams;Correlation;Local pattern;Pattern similarity
摘要:
Developing tools for monitoring the correlations among thousands of financial data streams in an online fashion can be interesting and useful work. We aimed to find highly correlative financial data streams in local patterns. A novel distance metric function slope duration distance (SDD) is proposed, which is compatible with the characteristics of actual financial data streams. Moreover, a model monitoring correlations among local patterns (MCALP) is presented, which dramatically decreases the computational cost using an algorithm quickly online segmenting and pruning (QONSP) with O(1) time cost at each time tick t, and our proposed new grid structure. Experimental results showed that MCALP provides an improvement of several orders of magnitude in performance relative to traditional naive linear scan techniques and maintains high precision. Furthermore, the model is incremental, parallelizable, and has a quick response time.
作者机构:
[李方敏; 刘新华] School of Information Engineering, Wuhan University of Technology, Wuhan 430070, China;[旷海兰] Computer Department, Hengyang Normal University, Hengyang 421008, China
通讯机构:
School of Information Engineering, Wuhan University of Technology, China
期刊:
Proceedings of the IEEE Power Engineering Society Transmission and Distribution Conference,2005年2005:1-7 ISSN:2160-8555
通讯作者:
Tong, X.(tongxj@csust.cn)
作者机构:
[Tong, Xiaojiao] Department of Mathematics, Changsha University of Science and Technology, Changsha 410076, China;[Tong, Xiaojiao] Institute of Mathematics, Changsha University of Science and Technology, China;[Tong, Xiaojiao] Changsha University of Science and Technology, China;[Lin, Mugang] Department of Computer, Hengyang Normal University, Hengyang 421008, China
通讯机构:
Changsha University of Science and Technology, China
关键词:
Decoupled method;KKT system;Optimal Power Flow;Semismooth Newton method
期刊:
PROCEEDINGS OF THE 2004 INTERNATIONAL CONFERENCE ON MACHINE LEARNING AND CYBERNETICS, VOLS 1-7,2004年3:1513-1518
通讯作者:
Jiang, SY
作者机构:
[Jiang, SY; Xu, YM] Hengyang Normal Univ, Dept Comp, Hengyang 421008, Peoples R China.
通讯机构:
[Jiang, SY] H;Hengyang Normal Univ, Dept Comp, Hengyang 421008, Peoples R China.
关键词:
distance;clustering;Data Mining
摘要:
A distance definition for mixed attribute and a simple method to calculate cluster parameter is proposed in this paper. Based on these, we present a clustering algorithm. The algorithm only scans over dataset one pass, and has the nearly linear time complexity with the size of dataset and the numbers of attributes, which make the algorithm deserve good scalability. Finally, we give empirical analysis to demonstrate the effectiveness, the experimental results show that the algorithm achieves both high quality clustering results and efficiency.