[1]朱 娟,吉根林.基于相交关系的GML空间线对象离群检测算法[J].南京师大学报(自然科学版),2010,33(03):127-130.
　Zhu Juan,Ji Genlin.An Algorithm for Detecting Outlier Lines Based on Intersection Relationship for GML Data[J].Journal of Nanjing Normal University(Natural Science Edition),2010,33(03):127-130.
点击复制

基于相交关系的GML空间线对象离群检测算法()
分享到：

《南京师大学报（自然科学版）》[ISSN:1001-4616/CN:32-1239/N]

卷:
第33卷
期数:
2010年03期
页码:
127-130
栏目:
计算机科学
出版日期:
2010-09-20

文章信息/Info

Title:
An Algorithm for Detecting Outlier Lines Based on Intersection Relationship for GML Data
作者:
朱 娟1;  吉根林2
1. 南京师范大学计算机科学与技术学院, 江苏南京210097 2. 江苏省信息安全保密技术工程研究中心, 江苏南京210097
Author(s):
Zhu Juan1; Ji Genlin2
1.School of Computer Science and Technology,Nanjing Normal University,Nanjing 210097,China 2. Jiangsu Research C enter of In form ation S ecu rity& Privacy Technology, N an jing 210097, C hina
关键词:
GML;  线对象;  相交关系;  离群检测
Keywords:
GML;  lines;  intersection re la tionsh ip;  outlier de tection
分类号:
TP311.13
摘要:
提出了一种基于相交关系的GML空间线对象离群检测算法DOL-IR,该算法首先计算GML线对象与其他空间对象的相交关系,定义基于相交关系的相异度,将其作为空间线对象之间距离的度量准则,利用DB-SCAN聚类算法检测离群的基于空间相交关系的线对象.实验结果表明,算法DOL-IR能准确地检测出离群的基于空间相交关系的线对象,并具有较高的效率.
Abstract:
A new a lgo rithm DOL- IR is presented for detecting outlier lines based on intersec tion re la tionsh ip for GML data. Inte rsection relations between spatia l lines and o ther spatia l ob jects a re compu ted. The d iffe rence degree betw een one line and another line is defined as the standa rd o f the d istance betw een one line and another line. A lgo rithm DBSCAN is used to detec t outlier lines based on intersection relationsh ip. The exper im enta l resu lts show that a lgo rithm DOL- IR can detec t outlie r lines based on intersection re la tionsh ip accurate ly and e ffectively.

参考文献/References:

[ 1] Barnett V, Lew is T. Outliers in S tatistical Data[M ]. New Yo rk: JohnW iley& Sons, 1994.
[ 2] Knorr E, Ng R. F ind ing intensional know ledg e o f distance-based outliers[ C ] / / Proc of the 25th Verg Larg e Databases Conference. Ed inburgh: M o rgan Kaufm ann Publishers, 1999: 211-222.
[ 3] BreunigM M, K rieg elH P, Ng R T, e t a.l Optics o:f identify ing density-based loca l outliers[ C] / / Proc of the 3rd European Con ference on Princ ip les and Practice o f Know ledge Discovery in Databases, Lecture No tes in Com puter Sc ience 1704. Prague: Spr inge r, 1999: 262-270.
[ 4] Prearata F, Sham osM. Compu tational Geom etry: An Introduc tion[M ]. Ber lin: Springer-Verlag, 1988.
[ 5] Jagad ish H V, Koudas N, M uthukr ishnan S. M in ing dev iants in a time series databases[ C ] / / Pro c o f the 25th Conference on Ve ry Large Databases. Edinburgh: M o rgan Kaufmann Pub lishe rs, 1999: 102-113.
[ 6] 郑建国, 焦李成. 偏差检测挖掘方法研究[ J]. 计算机工程, 2001, 27( 8): 33-35.
[ 7] Sheng Y ijiang, Q ing Boan. C lustering-based ou tlie r detec tion m ethod[ C ] / / 5 th Internationa l Conference on Fuzzy Sy stem s and Know ledge D iscovery. P iscataw ay: IEEE Com pute r Society, 2008: 429-433.
[ 8] 张书亮, 闾国年, 龚健雅, 等. 地理标示语言) ) ) Geo-W eb基础[M ] . 北京: 科学出版社, 2008: 3-4.
[ 9] 陈佳春, 吉根林. 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, 9( 1): 61-63.
[ 10] 李尼格, 鲍培明, 沙露. 一种基于面包含关系的GML空间离群面检测算法[ J]. 广西师范大学学报: 自然科学版, 2009, 27( 3): 118-121.
[ 11] 张宏, 温永宁, 刘爱利, 等. 地理信息系统算法基础[M ]. 北京: 科学出版社, 2006: 23-24.
[ 12] EsterM, K rieg elH P, Jorg S, et a .l A density-based a lgor ithm for d iscover ing cluster ing in large spa tia l da tabases w ith no ise[ C] / / Proceed ing o f 2nd Conference on Know ledge D iscov ery in Databases. Piscataway: IEEE Press, 1996: 226- 231.

备注/Memo

备注/Memo:
基金项目: 国家自然科学基金( 40871176) . 通讯联系人: 吉根林, 博士, 教授, 博士生导师, 研究方向: 数据挖掘技术及其应用. E-mail:glji@njnu. edu. cn.
更新日期/Last Update: 2013-04-08