Track: e-Business/e-Service/e-Commerce

Abstract

In e-commerce businesses, it is common nowadays to showcase products to the users on the basis of analytics which can turn those products into potential purchases. However, mostly e-commerce websites recommend only similar set of products on the basis of recent search history and not a curated set of products with higher confidence level of likeability. Rather than focusing just on a recent search history, if we can utilize social media data with this, it can open up a huge pool of data about a user and with this information, we will be able to provide well suited recommendations. Our objective in this paper is to develop an algorithm to provide very effective recommendations using the social media data of a user. This can help increase the efficiency of existing recommendation systems. Algorithm introduced in this paper follows a Fuzzy Logic technique to find relevant keywords emanated from Keywords Derivation algorithm which is a necessary algorithm for final result. It also makes use of Google Cloud Natural Language API to classify keywords in their respective categories. This algorithm is befitting to any products/services as long as the information about products and user’s social media data are available.