Product Recommendation Based on Content-based Filtering Using XGBoost Classifier

Abstract

Recommendation systems are a significant part of the machine learning algorithm to recommend related suggestions to the user based on user request. Many online shopping websites that have an acceptable rating information face problem in recommending an item to their users using content-based filtering (CBF) technique. By applying impermanent purchase patterns extracted from sequential pattern analysis (SPA) doesn’t make user satisfy from the search result. The objective of this research is XGBoost-based item recommendation using Jeju online shopping mall dataset records to recommend items based on user click information. Based on the output of XGBoost algorithm, we compare the result with other research outputs performance. The proposed CBF recommendation and SPA results successfully shows a better rating than other individual ones.