Analyzing Customer Buying Habits Using Transaction Data

Abstract

Data mining is the process of digging trough large files and databases to discover useful, non-obvious and often unexpected trends and relationships. Association rules are popular representations in data mining. It finds interesting association or correlation relationships among large set of data items. Most of the recent years, a very influential association rule mining algorithm, Apriori, has been used. It is to find frequent patterns, which produces candidate generation and multiple scans of database. Therefore, it is time consuming. Frequent pattern mining (FP-growth), is another milestone in development of association rules mining, which breaks the main bottlenecks of the Apriori. The frequent itemsets are generated with only two passes over the database and without any candidate generation process. This paper presents user buying habits using the sales transaction of stationery and FP-growth algorithm in association rule mining which is efficient and without candidate generation.