Dimension Reduction Techniques for Market Basket Analysis

Market basket analysis helps us to discover which group of items tends to be purchased together by customers. This is a powerful tool in data mining to understand the purchase behavior. Today the data sizes in the datasets of market basket have increased from gigabytes to terabytes or even larger due to which the complexity of analysis of huge datasets has been a major concern in almost all areas of technology in the past decade. A solution to this crucial problem in distributed data mining is that, the massive dataset can be collected and warehoused in a single site if its dimensionality is reduced. Today many dimension reduction algorithms are there and they are generally classified into feature selection, feature extraction and random projection. In this paper a new dimension reduction algorithm is proposed, which is different from all the existing methods, to encode the transactions which reduces the Size of transaction that in turn reduces the communication cost and communication bandwidth.