Efficient Mining of Top-K Strongly Correlated Item Pairs using One Pass Technique

This paper presents an efficient one pass technique, k-SCOPE (Top k Strongly Correlated item Pair Extraction), which finds top-k strongly correlated item pairs from transaction database, without generating any candidate sets. The proposed technique uses a correlogram matrix based approach to compute support count of all the 1- and 2-itemset in a single scan over the database. From the correlogram matrix the correlation values of all the item pairs are computed and top-k correlated pairs are extracted very easily. The simplified logic structure makes the implementation of the proposed technique more attractive. We experimented with real and synthetic datasets and compared the performance of the proposed technique with its other counterparts and found satisfactory.