ISSN : 2319-7323

INTERNATIONAL JOURNAL OF COMPUTER SCIENCE ENGINEERING

ABSTRACT

Title	:	Association Mining Approach for Customer Behavior Analytics
Authors	:	D.M.R.M Dissanayake, S. C. Premaratne
Keywords	:	Data mining, FP-Growth Algorithm, Frequent Item set Mining, K-means cluster analysis, Consumer behavior, Association Rules
Issue Date	:	Jan-Feb 2020
Abstract	:	This research suggests a proper decision-making system using Data mining technique such as Frequent Pattern Growth (FP-Growth) Algorithm and K-means clustering, in order to identify consumer behavior on trending food items and conduct profitable marketing campaigns and promotions by comparing association rules of a particular date or day with the previous year. Once the system is developed, timely promotion creation can be done in a more consistent and straight forward way rather than promoting items in a senseless way by comparing the previous year�s same season association behaviors. In order to create such promotions, the proposed system contains data which is undergone through data mining Algorithms. In developing the system, past transaction data is collected from the Point of sales system, and data preprocessing is done by data mining preprocessing techniques. One of the main activities in a food outlet is to determine associations, the inherent regularities in data such as products purchased together and what is the likelihood of buying a specific product after purchasing a certain product. Niwa Sushi Pte Ltd, Singapore as a Japanese food outlet has not yet been used such categorization and consideration in their sales system for processing associations, frequent itemset and subsequent items. The process is still not done even manually, but occasionally done by random observation & heuristic on sales data. The process conducted manually is also not the most accurate information but kind of near guesses.
Page(s)	:	14-39
ISSN	:	2319-7323
Source	:	Vol. 9, No. 1