Collaborative filtering based online recommendation systems: A survey

Abstract

In recent years, the volume of data present online has grown exponentially. A major portion of this data is related to internet-based e-commerce platforms. The evaluation of such data and/or the extraction of information is difficult due to its huge volume. It is cumbersome for an individual or an organization to obtain the desired results in a timely manner. Recommender Systems (RS) present an automated and efficient solution to this problem. Recommender systems analyze the user profile/behavior and presents products relative to the users' interests. RS maybe based on collaborative filtering, content-based or a hybrid of these techniques. Online recommendation through Collaborative Filtering (CF) plays a vital role in e-commerce and is regarded as one of the best techniques for making possible recommendations for customers. This research analyzes the recommendation systems based on collaborative filtering. Two techniques applied in recommendation system based on collaborative filtering are item-based and user-based approaches. In todays' world, these techniques take place in a global internet environment to produce accurate results according to the need of the user. This paper presents a survey of various state of the art techniques for recommendation systems and highlights the best techniques to generate accurate results.