Proceedings of the IEEK Conference (대한전자공학회:학술대회논문집)
- 2003.11b
- /
- Pages.207-210
- /
- 2003
Discovery of Association Rules Base on Data of Time Series and Quantitative Attribute
시간적 관계와 수량적 가중치 따른 연관규칙 발견
Abstract
In this paper, we explore a new data mining capability that is based on Quantitative Attribute and Time Series. Our solution procedure consists of two steps. First, We derive an algorithm to contain the Quantitative Attribute into a set of candidate item. Second, We redefine the concepts of confidence and support for composite association rules. It is shown that proposed methode is very advantageous and can lead to prominent performance improvement.
Keywords