Thursday 22nd of February 2018

Customer Buying Behavior Analysis: A Clustered Closed Frequent Itemsets for Transactional Databases

N.Kavitha and S.Karthikeyan

The problem in data mining applications is the mining of frequent patterns. Though there has been various techniques, such as pattern discovery, association rule mining etc, these methods generates a large volume of frequent patterns and rules which are not useful for finding the essential patterns among them, from the database. Alternatively, CFIM technique produces a relatively lesser number of closed frequent patterns. Patterns are pruned before clustering and the clustered patterns help in predicting the customer purchasing behavior which in turn helps to maintain an inventory and focus the point of sale on transaction data, enhancing sales. This can be achieved by CF-CLUS algorithm which clusters the similar patterns from the generated closed frequent patterns. The distinguishing feature of CF-CLUS is the ability to reduce the number of scans of the database and works well when the minimum support is low and produce good results if the database is sparse. This led to efficient clustering of the results from the generated closed patterns.

Keywords: Association Rule Mining, Pattern Discovery, Cluster Patterns and Market Basket Data, CFIM-Closed frequent itemset mining.

Download Full-Text


Research Scholar,Departmet of Computer Scence,Karpagam University

Assitant Professor,Department of Information Technology, College of Applied Sciences, Oman

IJCSI Published Papers Indexed By:





IJCSI is a refereed open access international journal for scientific papers dealing in all areas of computer science research...

Learn more »
Join Us

Read the most frequently asked questions about IJCSI.

Frequently Asked Questions (FAQs) »
Get in touch

Phone: +230 911 5482

More contact details »