| 基于引力的孤立点检测算法 |
| |
| 引用本文： | 孟建良,姚亮,程伟想.基于引力的孤立点检测算法[J].计算机应用与软件,2009,26(1). |
| |
| 作者姓名： | 孟建良 姚亮 程伟想 |
| |
| 作者单位： | 华北电力大学计算机学院,河北,保定,071003 |
| |
| 摘 要： | 提出一种基于引力的孤立点检测算法.通过综合考虑数据对象周围的密度及数据对象之间的距离等因素对孤立点定义的影响来挖掘出数据集中隐含的孤立点.给出了与该算法相关的概念与技术,详细介绍了该算法,并用实际数据进行了实验.实验表明:该算法对数据集的维度具有很好的扩展性,能有效地识别孤立点,同时能反映出数据对象在数据集中的孤立程度.
|
| 关 键 词： | 孤立点检测 孤立点 引力 |
ALGORITHM OF GRAVITY-BASED OUTLIERS DETECTION |
| |
| Meng Jianliang,Yao Liang,Cheng Weixiang.ALGORITHM OF GRAVITY-BASED OUTLIERS DETECTION[J].Computer Applications and Software,2009,26(1). |
| |
| Authors: | Meng Jianliang Yao Liang Cheng Weixiang |
| |
| Affiliation: | School of Computer;North China Electric Power University;Baoding 071003;Hebei;China |
| |
| Abstract: | A gravity-based outliers detection algorithm(GODA) is presented.Because of both the density around a data object and the distance between it and the others can influence outlier's definition,by comprehensively considering these two parts,the algorithm can detect the crytic outliers in the dataset.In this paper it proposes the definitions and techniques with regard to the algorithm and introduces the algorithm in detail.Experiments have been carried out with actual data,and the results indicate that the new ... |
| |
| Keywords: | Outlier detection Outlier Gravity |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
