Efficient algorithm for the extraction of association rules in data mining

  • Authors:

  • Pinaki Mitra;Chitrita Chaudhuri

  • Affiliations:

  • Indian Institute of Technology Guwahati, North Guwahati, Guwahati, India;Department of Computer Science and Engineering, Jadavpur University, Kolkata, India

  • Venue:
  • ICCSA'06 Proceedings of the 2006 international conference on Computational Science and Its Applications - Volume Part II
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

The problem of data mining is to discover the pattern or trend in huge volume of data. The problem is similar to knowledge discovery in artificial intelligence. Here our goal is to discover rules that reflect the pattern in the data. These rules are called association rules. In [AS94] an algorithm is proposed to extract these association rules from the large/frequent itemsets computed by the apriori algorithm. In this paper we present a more efficient and output sensitive algorithm to compute these association rules given the lattice L of large itemsets. Our approach is based on pruning a lot of redundant association rules that have to be tested in the algorithm of [AS94] .We use a variation of the data structure for hashing using separate chaining in our algorithm. Our algorithm, is output sensitive in the sense that its time complexity will be proportional to the number of association rules that have to be generated and it is also optimal.