Efficient association rule mining for market basket analysis
Data mining is an attitude that business actions should be based on learning, that informed decisions are better than uninformed decisions, and that measuring results is beneficial to the business. Data mining is also a process and a methodology for applying the tools and techniques. Association rule mining is also one among the most commonly used techniques in Data mining. A typical and the most running example of association rule mining is market basket analysis. This process analyzes customer buying habits by finding associations between the different items that customers place in their “shopping baskets”. The discovery of such associations can help retailers develop marketing strategies by gaining insight into which items are frequently purchased together by customer and which items bring them better profits when placed with in close proximity.
The algorithms for single dimensional association rule mining, such as apriori and the FPtree developed are in a greater use today. However, candidate set generation in apriori is still costly, especially when there exists a large number of patterns and/or long patterns. And both these algorithms prune the itemsets based on their frequencies (i.e., if their frequencies exceed minimum support threshold then they term them as frequent and the rest of them as infrequent). But this pruning technique is insufficient to help market analyst to make decisions such as planning the supermarket's shelf space, changing the layout new store layouts, new product assortments, which products to put on promotion so as to improve their marketing profits. So the focus of this thesis is to enhance these algorithms in a way that it provides frequent profitable patterns which help market analyst to make the best informed decisions for improving their business.
Efficient Association, Rule Mining, Market Basket Analysis.