Quantitative Association Rules Based on Distance

In association analysis, mining the continuous attributes may reveal useful and interesting insights about the data objects which are of continuous attributes. Quantitative association rules are aimed to deal with the relationships among continuous attributes of data objects. This paper presents an association analysis algorithm based on the distances among clusters. The algorithm uses a clustering algorithm to identify the intervals of attributes in clusters and combines the clusters projected on attributes to form distance-based association rules. Experimental analysis indicates that the algorithm is effective in real world applications.