A review on improving recommendation quality by using relevant contextual information
Sanjay Kumar Dwivedi, Bhupesh Rawat · International Conference on Computing for Sustainable Global Development · 2016
As huge amount of data is being generated on the web rapidly, so it is becoming extremely difficult for the users to find the most relevant items. To address this problem recommender systems have been used for a long time to recommend items to the users from a large repository based on their interest and preferences. However many of the existing recommender systems do not consider the contextual information including time, location etc which have an important part in improving the quality of recommended items. The benefits of using contextual information have been confirmed by research community and practitioners as well in several areas of computer science. This paper discusses the notion of context, then goes on explaining how to model contextual information in recommender systems, reviews various methods of obtaining contextual information, describes different paradigms of including context in various stages of recommendation process depending upon the requirement of application, then it discusses the critical issues in context aware recommendation systems and finally suggests approaches that have not been explored yet and could help in improving recommendation quality.