In this paper we look into emotion based recommender systems, the approach that researcher have used to build an emotion based recommender system. We review the method used for emotion detection first followed by the method used for building the recommender system. While emotion recommender systems are not very widely used at the moment we believe as we are moving towards a world of automation where most human activities are being automated, emotion based recommender systems have a great role to play. Fields such as intelligent systems and internet of things will benefit greatly with the concept of emotion based recommender system. The concepts of computer vision, sentiment analysis, medical forms of emotion detection are all applicable therefore providing the concept of emotion based recommender systems with a wide variety of promising research areas. The medical world is also not left behind as such automated systems can greatly be build to serve the mental health and non-verbal communities