基于DS理论的不确定空间co-location模式挖掘 |
| |
作者单位: | 云南大学 信息学院 计算机科学与工程系 |
| |
基金项目: | 云南大学科学研究基金资助项目(2009F29Q);国家自然科学基金资助项目(61063008);云南省自然科学基金资助项目(2010cd025). |
| |
摘 要: | 提出一种基于DS理论的co-location挖掘方法.先将数据集按照空间对象分类,然后从分类后的子数据集中依次提取该空间对象的实例,与其他空间对象数据集中的各个实例进行配对,将找到的各实例关系记录到一个新的关系表中,作为使用DS理论求解的基础.从一个新的角度来研究co-location挖掘技术,在DS理论的基础上重新定义了参与度,基于新定义的参与度来求解co-location模式.
|
关 键 词: | 数据挖掘 不确定性类型 DS证据理论 同位模式 |
Mining co-location patterns over uncertain data based on DS evidence theory |
| |
Affiliation: | Department of Computer Science and Engineering,School of Information Science and Engineering,Yunnan University |
| |
Abstract: | This paper proposed a method of co-location pattern mining over uncertain data based on DS evidence theory.Firstly,the spatial objects were classified into data subsets.Then,selecting the instances of a spatial object from the classified subsets successively matched each instance in the rest of spatial object subsets.Finally,for using DS theory to solve co-location pattern mining,the found instance relations were recorded into a new table.In this paper,a new perspective to study the co-location mining.Based on the DS theory,a new concept of the participation index was defined,and a co-location pattern mining algorithm was designed based on the new definition. |
| |
Keywords: | spatial data mining uncertain data DS evidence theory co-location pattern |
|
| 点击此处可从《云南大学学报(自然科学版)》浏览原始摘要信息 |
|
点击此处可从《云南大学学报(自然科学版)》下载全文 |