| 基于Voronoi和空间自相关的离群点检测 |
| |
| 引用本文： | 王 妍,潘瑜春,阎波杰. 基于Voronoi和空间自相关的离群点检测[J]. 计算机工程, 2010, 36(1): 33-34,3 |
| |
| 作者姓名： | 王 妍 潘瑜春 阎波杰 |
| |
| 作者单位： | (1. 首都师范大学信息工程学院，北京 100037；2. 国家农业信息化工程技术研究中心，北京 100097；3. 闽江学院地理科学系，福州 350108) |
| |
| 摘 要： | 为了提高空间数据挖掘的效率和准确度，在分析传统的离群点检测算法优、缺点的基础上，提出一种空间离群点检测算法。用Voronoi来确定空间对象间的邻近关系，在空间邻域内利用空间自相关性来计算局部Moran指数，并将其作为离群因子进而判断离群点。实验结果表明，该算法能够高效、准确地检测出空间离群点，具有对用户依赖性少和可伸缩性强等优点。
|
| 关 键 词： | 空间离群点 Moran指数 空间自相关 |
Outlier Detection Based on Voronoi and Spatial Autocorrelation |
| |
| WANG Yan,,PAN Yu-chun,YAN Bo-jie,. Outlier Detection Based on Voronoi and Spatial Autocorrelation[J]. Computer Engineering, 2010, 36(1): 33-34,3 |
| |
| Authors: | WANG Yan PAN Yu-chun YAN Bo-jie |
| |
| Affiliation: | (1. Computing Center, Northeastern University, Shenyang 110004; 2. College of Information Science and Engineering, Northeastern University, Shenyang 110004) |
| |
| Abstract: | Mobile Element(ME) are usually exploited for collecting and relaying data in sparse wireless sensor networks. This paper proposes the Voronoi Diagram-based Mobile Element Schedule(VDMES) algorithm to construct the shortest possible path for ME data collection. ME are scheduled to visit a small subset of Voronoi vertices, which exactly covers all sensor nodes in a given transmission radius. Simulation experimental result shows that the path concatenating the Voronoi vertices is much shorter than that formed by regular sensor nodes. |
| |
| Keywords: | sparse wireless sensor networks Voronoi diagram Mobile Element(ME) data collection |
|
| 点击此处可从《计算机工程》浏览原始摘要信息 |
|
点击此处可从《计算机工程》下载免费的PDF全文 |
