Association rule mining is one of the most important problems of data mining. Since remotely sensed data contains huge amounts of information, it’s a very potent area to discover useful rules. Compared to traditional Market “basket data”, remote sensed imagery has specific characteristics and also presents specific difficulties. Two problems need to be solved to apply association rule mining on remotely sensed images. The first is to deal with quantitative attributes. The second is to efficiently handle huge quantities of information. For the first problem, partitioning quantitative data into intervals is a simple but effective way. For the second problem, we propose a new approach based on transaction patterns and occurrence counting, which simplifies the calculation of support and is much more efficient. A modified Apriori Algorithm is given for which performance analysis shows obvious improvements.
[1]
Rakesh Agarwal,et al.
Fast Algorithms for Mining Association Rules
,
1994,
VLDB 1994.
[2]
Philip S. Yu,et al.
An effective hash-based algorithm for mining association rules
,
1995,
SIGMOD '95.
[3]
Ramakrishnan Srikant,et al.
Mining quantitative association rules in large relational tables
,
1996,
SIGMOD '96.
[4]
Renée J. Miller,et al.
Association rules over interval data
,
1997,
SIGMOD '97.
[5]
Jianning Dong,et al.
The application of association rule mining to remotely sensed data
,
2000,
SAC '00.
[6]
Tomasz Imielinski,et al.
Mining association rules between sets of items in large databases
,
1993,
SIGMOD Conference.