Exploring Spatial ARM (Spatial Association Rule Mining) for Geo-Decision Support System

Ranjana Vyas,Lokesh Kumar Sharma,U.S Tiwary

doi:10.3844/jcssp.2007.882.886

Abstract

Geographical Decision Support System (Geo-DSS) is a demanding field, since enormous amount of spatial data have been collected in various applications, ranging form Remote Sensing to GIS, Computer Cartography, Environmental Assessment and Planning. Although some efforts were made to combine spatial mining with Spatial Decision Support System but mostly researchers for spatial database are using a popular data mining approach-Apriori based association rule mining. There are two major limitations in existing approaches; the biggest being, that in a typical Apriori based spatial association the same records are required to be scanned again and again to find out the frequent sets. This becomes cumbersome, as spatial data is already known to be large in size. As far as sparse data is concerned, an Apriori based spatial association rule may even be considered but when there is dense data there were other approaches giving better performance. Researchers discuss only the positive spatial association rules; they have not considered the spatial negative association rules. Negative association rules are very useful in some spatial problems and are capable of extracting some useful and previously unknown hidden information. As this approach makes computation faster, it is thus better candidate for integration into Geo-DSS architectural framework. We have tried to design a particular Decision support system using spatial positive and negative association rule with efficient P- Tree and T-Tree.

Highlights

Decision Support Systems have been under discussion, development and investigation by Information Systems researchers for more than 35 years
In this study we have proposed a novel approach of mining spatial positive and spatial negative association rule mining using P tree and T tree which are very useful in some spatial problems and capable of extracting some useful and previously unknown hidden information
There are a number of features of the P-tree Table that enhance the efficiency of this process: large K-predicate sets can be extracted at the sec level and again the spatial Association rule can be extracted from it

Summary

INTRODUCTION

Decision Support Systems have been under discussion, development and investigation by Information Systems researchers for more than 35 years. J. Computer Sci., 3 (11): 882-886, 2007 major limitations and the biggest disadvantage of Apriori based spatial Association rule is that same datasets are required to be scanned again and again in order to find out the frequent sets. Computer Sci., 3 (11): 882-886, 2007 major limitations and the biggest disadvantage of Apriori based spatial Association rule is that same datasets are required to be scanned again and again in order to find out the frequent sets This approach is very cumber some in general and it becomes more expensive with spatial data, which is known to be large sized. In this study we have proposed a novel approach of mining spatial positive and spatial negative association rule mining using P tree and T tree which are very useful in some spatial problems and capable of extracting some useful and previously unknown hidden information. A conjunction of k single predicates is called a k-predicate

INCORPORATING INTERESTING ITEM SET

Association rule discovery seeks rules of the form

Of entities is computed at a relatively coarse

CONCLUSION