A recommendation system is that type of system which uses the entities such as “likes”, “preferences” and “ratings” in order to predict the items that a user would want. It is kind of a filtering system which suggests the users what products they would like to see. This technique can be used to counter the information overloading. In the current online world, information overloading is a major issue as we have a plethora of choices and this can be done using personalized systems. The recommendation systems can mainly classified into three types. The first is based on the content where the system uses the information regarding the other user to suggest and recommend the items which are most likely to satisfy the customer needs. Next is the collaborative filtering, where the own information of the individual is only taken into consideration. Finally, comes the trust-aware system, where the information related to the social media and its trust information is used to predict the user’s likings and his preferences.K- nearest neighbour(k-NN) is one of the best algorithms when it comes to the collaborative filtering(CF) and is used to fine tune the recommendation systems. It also deals with the information overloading problem which is mentioned above by generating predicted ratings for all those users who have not expressed their opinions. Here, in this paper we will try to discuss how one can use the trust as an alternative in order to overcome the limitations of collaborative filtering. A k-nearest recommenders (k-NR) algorithm is proposed in the paper where the user learns who all he or she should trust and up to what extent he can trust, by studying and assessing the ratings that the individuals have received. © Blue Eyes Intelligence Engineering & Sciences Publication.