Personalized web recommendations analyzing sequential behaviour using implicit data streams: A survey
Shalini Gupta, Uma Ojha, Veer Sain Dixit · 2017
The ongoing expansion of commercial websites greatly increases the need of effective recommender systems for finding the solution to information overload problem. Conventional systems made use of explicit information obtained from users in the form of ratings and feedback queries which are highly affected by users' demographic location, age and attitude. The methodology used nowadays is to make use of implicit information of users to remove this constraint. This implicit data can be users surfing behavior, personal information or contextual situations that can be used to improve the accuracy of recommendations. This article provides a review on recent developments in making recommendations based on data that is extracted by monitoring the surfing behavior of customers and applying various information filtering approaches. The domain used in the study is click stream data of e-commerce website. We compare and evaluate available algorithms and examine their role in future developments.