Customer Review Based Product Recommendation in E Commerce Using Improved Frequent Pattern Mining and Artificial Intelligence
Loading...
Date
item.page.authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Recommendation plays a very vital role in human life. Human beings rely a lot on
newlinerecommendations from their daily routines to taking any big decision i.e. purchasing
newlinenew things, organizing a function, recruiting a resource, buying furniture, or a new
newlinehome. People reliance on recommendations and take their decisions based on the
newlinerecommendation received from various sources. In our research work, we are focusing
newlineon providing recommendations to the customers of e-commerce web sites. As the title
newlinesuggests our recommendation system is based on customers reviews. While shopping
newlinethrough an e-commerce website, if the customer gets confused in selecting a product
newlineout of many available options then the e-commerce platform provides a comparison
newlineoption based on the features of the product, but what if the user can read the reviews of
newlinethe product and then can decide that with which product he/she should go for purchase.
newlineFurther, it is very much difficult or we can say next to impossible for the customers to
newlinewalk through the thousands of available reviews of any single product and then compare
newlinewith other product to decide that with which product he/she should proceed for
newlinepurchase. To recommend a product based on reviews we need to deal with the text and
newlinehence improved frequent pattern mining has been implemented to extract the relevant
newlinecontent and finally artificial intelligence applied to the extracted relevant content. We
newlinehave also tested our output in a machine learning algorithm named Random Tree to
newlinevalidate our developed algorithm. Our proposed framework is divided into four phases,
newlinewhich include phase 1 - Products on the e-commerce website. Phase 2 - Users reviews
newlineand ratings. Phase 3 - Improved frequent pattern mining. And in final Phase 4 - Artificial Intelligence has been implemented.
newline