Weak Ratio Rules: A Generalized Boolean Association Rules

  • Authors:
  • Baoqing Jiang;Xiaohua Hu;Qing Wei;Jingjing Song;Chong Han;Meng Liang

  • Affiliations:
  • Henan University, China;Henan University China, and Drexel Univeristy, USA;Henan University of Economics and Law, China;Qingyuan Polytechnic, China;Henan University, China;Henan University, China

  • Venue:
  • International Journal of Data Warehousing and Mining
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper examines the problem of weak ratio rules between nonnegative real-valued data in a transactional database. The weak ratio rule is a weaker form than Flip Korn's ratio rule. After analyzing the mathematical model of weak ratio rules problem, the authors conclude that it is a generalization of Boolean association rules problem and every weak ratio rule is supported by a Boolean association rule. Following the properties of weak ratio rules, the authors propose an algorithm for mining an important subset of weak ratio rules and construct a weak ratio rule uncertainty reasoning method. An example is given to show how to apply weak ratio rules to reconstruct lost data, and forecast and detect outliers.