[1]陈佳春,等.基于空间相邻关系的GML点对象离群检测算法[J].南京师范大学学报(工程技术版),2009,09(01):061-63.
　Chen Jiachun,Ji Genlin,et al.An Algorithm for GML Point Outlier Detection Based on Space Adjacent Relations[J].Journal of Nanjing Normal University(Engineering and Technology),2009,09(01):061-63.
点击复制

基于空间相邻关系的GML点对象离群检测算法
分享到：

南京师范大学学报（工程技术版）[ISSN:1006-6977/CN:61-1281/TN]

卷:
09卷
期数:
2009年01期
页码:
061-63
栏目:
出版日期:
2009-03-30

文章信息/Info

Title:
An Algorithm for GML Point Outlier Detection Based on Space Adjacent Relations
作者:
陈佳春1;  2;  吉根林1;  2
1. 南京师范大学数学与计算机科学学院, 江苏南京210097; 2. 南京师范大学虚拟地理环境教育部重点实验室, 江苏南京210097
Author(s):
Chen Jiachun1; 2; Ji Genlin1; 2
1.School of Mathematics and Computer Science,Nanjing Normal University,Nanjing 210097,China;2.Key Laboratory of Virtual Geographic Environment,Ministry of Education,Nanjing Normal University,Nanjing 210097,China
关键词:
离群点检测;  空间相邻;  GML数据挖掘
Keywords:
outlier detec tion;  space ad jacent re lations;  GML da tam in ing
分类号:
TP311.13
摘要:
提出了一种基于空间相邻关系的点对象离群检测算法SAOD(Space Adjacent Relations Based GML Point Outlier Detec-tion Algorithm).利用空间相邻关系作为空间点对象的相似度度量准则,得到相似度矩阵,从而挖掘GML中的离群点对象.实验结果表明,SAOD算法能有效地检测GML中的离群点对象并且具有较高的效率.
Abstract:
A t present algorithm fo rGM L outlier de tection has se ldom been researched. A lgor ithm SAOD for GML po int outlier de tection based on space ad jacent re la tions is propo sed in th is paper. In th is a lgor ithm, the space adjacen t relations be tw een spatia l po ints are considered as the sim ilar itym easurem ent and sim ilar itym atrix is computed. The expected outliers can be ob tained from the m atr ix. The resu lts o f experim ents show that a lgor ithm SAOD is effec tive and eff-i c ient.

参考文献/References:

[ 1] Barnett V, Lew is T. Outliers in S tatistical Data[M ]. New Yo rk: JohnW iley& Sons, 1994: 194-223.
[ 2] Knorr EM, Ng R T. A lgo rithm s fo rm in ing distance-based outliers in larg e da tasets[ C] / /Proc of 1998 Internationa l Conference. Very Large Da ta Base ( VLDB98) . New Yo rk: VLDB Endowm ent, 1998: 392-403.
[ 3] BreunigM M, KruegerH P, Ng R, et a.l LOF: identify ing density-based local outliers[ C ] / /Proc o f ACM SIGMOD2000 Internationa l Con ference onM anagem ent of Da ta. Da llas, Texas: ACM Press, 2000: 93-104.
[ 4] Knorr EM, Ng R T. A un ified notion o f outliers: properties and computation[ C] / /Proc the 3rd Inte rnational Confe rence on Know ledg e Discovery and Da taM ining. Ca lifo rnia: IEEE Press, 1997: 219-222.

备注/Memo

备注/Memo:
基金项目: 国家自然科学基金( 40771163, 40871176)资助项目.
通讯联系人: 吉根林, 教授, 博士生导师, 研究方向: 数据挖掘技术及应用技术、XML技术. E-mail:glji@n jnu. edu. cn
更新日期/Last Update: 2013-04-23