DOI: 10.7763/IJCTE.2012.V4.620
Data Partitioning and Bit Vector Approach for Weighted Frequent Item Set Mining
Abstract—Association rule mining is an important task in data mining. Association rules are frequently used by retail stores to assist in marketing, advertising, floor placement and inventory control. Most of the association rule mining algorithms will not consider the weight of an item. Weighted association is very important in KDD.In analyzing market basket analysis, people often use apriori algorithm, but apriori generates large number of frequent item sets. One alternate approach to apriori is partitioning technique. This paper presents a method to find weighted frequent item sets using partitioning and bit vector .By example, it is proved that partitioning technique can improve the efficiency by reducing the number of candidates.
Index Terms—Weighted association rules, KDD, partitioning, bit vector, weighted support.
M. Jabbar is with the JNTU Hyderabad, India (e-mail: Jabbar.meerja@gmail.com).
B. L. Deekshatulu is with the Visiting Professor HCU, Hyderabad, India.
P. Chandra is with the Scientist Advanced systems Laboratory India.
Cite: M. A. Ja bbar, B. L. D eekshatulu, and Priti Chandra, "Data Partitioning and Bit Vector Approach for Weighted Frequent Item Set Mining," International Journal of Computer Theory and Engineering vol. 4, no. 6, pp. 980-982, 2012.