节点文献
带正则化项的时间序列聚类算法及其应用
Clustering Method with Regularization for Time Series Data and Its Application
【摘要】 为了更加准确地对时间序列数据进行聚类分析,运用了带正则化项的时间序列聚类方法,并实现了该聚类方法的算法.将该方法应用于云南地区水准形变数据的实际研究中,以寻找类之间状态转移与地震的关系.数值结果同时表明了带正则化项的时间序列聚类方法比标准的K-means方法更有效,更有优势.
【Abstract】 In order to have a better clustering result of the time series data,a new time series clustering method with regularization is introduced.Applying this method into studying the horizontal deformation data of Yunnan province,the relationship between the transition of the states and the eruption of the earthquakes could be figured out.The numerical results show that,comparing with the standard K-means method,this approach manifests its efficiency.
【关键词】 正则化;
时间序列聚类方法;
标准K-means方法;
水准形变;
【Key words】 regularization; time series clustering method; standard K-means method; horizontal deformation;
【Key words】 regularization; time series clustering method; standard K-means method; horizontal deformation;
- 【文献出处】 复旦学报(自然科学版) ,Journal of Fudan University(Natural Science) , 编辑部邮箱 ,2012年04期
- 【分类号】O212.1
- 【被引频次】1
- 【下载频次】99