Abstract:
Many online communities including the popular social networks offer automated recommendations to their users. This automated recommendations are normally generated using collaborative filtering systems based on the past ratings or opinions of the similar users. Alternatively, trust among the users in the network also can be used to find the neighbors while making recommendations. To obtain the optimum result, there must be a positive correlation exits between trust and interest similarity. In this paper we have presented the result of the survey on the relationship between trust and interest similarity. Our result supports the assumed hypothesis of positive relationship between the trust and interest similarity of the users.
