Product recommendation for the day using fuzzy c-means and association rule generator in KNIME

Online retail stores will need to explore various marketing strategy to attract customers. Each customer would buy a varying number of products, from the categories available. They also check for offers and recommendations before placing the orders. The design proposed is suitable for small to medium retailers. In this paper, a recommendation system is developed using KNIME tool by employing the data mining techniques like preprocessing, clustering, frequent item set detection and then use the classification oriented cluster analysis measures: the Precision, Recall, F-measure and Accuracy for verifying the results of recommendations.

I. Introduction

Internet is reaching everywhere. These days the small and medium retail stores too host their own websites for the daily needs of the customers and they even provide delivery services for the orders in their locality. Customers are getting accustomed of this and more over order arrives at their door step in couple of hours. While this is happening the retailer needs to market his products and also do recommendations through his website.