Author : Geeta V. Poul and P.D. Lambhate

Data mining is used to find interesting information from the raw data. The frequent itemset mining used to find number of itemsets occurred more number of times in particular time duration. It may happen that a particular item occurs for a very specific time but its frequency is more. Such itemsets are considered as non-redundant itemsets. Thus the study of temporal data mining (change mining) is important. Number of data mining algorithms introduced to find frequent itemsets in the data. The work is based on HIGEN miner algorithm to find redundant as well as non-redundant itemsets. The proposed work finds HIStory GENeralized pattern (HIGEN) using automatic inference taxonomy in a very less time. The experiment performed on both synthetic and real time datasets to find value satisfying minimum support at higher level of taxonomy.

Keywords: Data mining; Minimum support; Association rule; change mining.

Article published in International Journal of Current Engineering and Technology, Vol.5, No.4 (Aug-2015)