AN ADAPTIVE TRANSACTION REDUCTION APPROACH FOR MINING FREQUENT ITEMSETS: A COMPARATIVE STUDY ON DENGUE VIRUS TYPE1
K. P. Kaliyamurthie · International Journal of Pharma and Bio Sciences · 2015
Frequent itemset mining plays an essential role in mining various patterns and in real time applications. The dataset utilised in our experimental analysis are real world data set for Dengue Virus Type 1 (DEN1) which is obtained from GenBank:AAB27904.1 which consists of 777 amino acids.In this paper, an adaptive TDTR(Two Dimensional Transactions Reduction) approach which we have proposed earlier is tested against this real Dengue virus type1 dataset and finally compared with standard Apriori algorithm and FP-Growth algorithm. The theoretical analysis and experiments prove its efficiency and accuracy for Dengue Virus Type1 dataset . This system reveals that Leucine(L), Phenylalanine (F),Lysine(K),Serine(S) and Glycine(G) are the dominating amino acids in Dengue Virus Type1 which is the same results produced from Apriori algorithm and FP- Growth Algorithm with high performance.