An intelligent recommendation system in e-commerce using ensemble learning

In today's world, recommendation systems play a vital role in customer analysis on social media, online businesses, e-commerce, etc. There are multiple sources of information on the Internet giving people a large set of suggestions and advice. This may create confusion for the accurate decision to the user and he/she may get lost in the competitive and growing market. A recommendation system is an essential part of e-commerce to supply the filtered relevant information asked by the customer. The major pitfalls of the existing recommendation system are flooding unnecessary recommendations and unpredictability about new products. Most of the recommendation systems rely on the purchase history of the customer and give suggestions for new products. Along with the history of the user's purchase, it is crucial to analyze various other activities such as browsing history, wish lists, reviews, ratings, and previously ordered items. An intelligent recommendation system using ensemble learning is presented in this paper to reduce duplicate and irrelevant recommendations for the customer. The experimental results indicate that there has been a significant improvement in the precision and recall of the recommendation system in comparison with the other conventional techniques.