A Hybrid Recommendation System Based on Collaborative Filtering and Association Rule

Authors: Khushboo Sahu, Prof. Mayank Bhatt

Certificate: View Certificate

Abstract

A recommendation engine is a feature (not a product) that filters items by predicting how a user might rate them. It solves the problem of connecting your existing users with the right items in your massive inventory (i.e. tens of thousands to millions) of products or content. Product recommendations are a must-have feature for all ecommerce websites, as they can drive sales, increase conversion rate and order value. Sending personalized product recommendations to your customers increases sales in just a few clicks. Add a Product Recommendations content block or merge tag to E-commerce Automation workflows to encourage subscribers to visit your store, re-engage inactive customers by promoting relevant items, suggest your best-selling products to new customers, and much more. Most of web portals have integrated this feature but web personalization is still a demanding issue. This project works target to integrate web personalization feature with product recommendation to enhance the sales and performance of E-commerce portal. A hybrid recommendation algorithm based on collaborative filtering and association rule will be used to recommend the products. Java technology will be used to develop these solutions.

Introduction

The growing internet world and hectic schedule of daily life create so much difficulty for internet Users to find desired information. This situation becomes worse when user try to search information and get irrelevant information. Inadequate knowledge of search tool and large amount of data gives poor performance to retrieve or extract desire information. Recommendation systems offer intellectual practice based on user preference. Recommendation systems offer separate and specialized set of information. In recent years, Web personalization has received much attention to help Internet users with the problem of information overload.

Conclusion

The complete work expects an integrated solution based on product characteristics and customer behavior for product recommendation. Few points to describe the expectations are illustrated below; 1. To develop customer classification algorithm for customer behavior analysis. 2. To calculate product similarity and popularity index to estimate sales trend and popular product list. 3. Product recommendation based on product and customer nature.