Using the opinion leaders in social networks to improve the cold start challenge in recommender systems

The increasing volume of information about goods and services has been growing confusion for online buyers in cyberspace and this problem still continues. One of the most important ways to deal with the information overload is using a system called recommender system. The task of a recommender system is to offer the most appropriate and the nearest product to the user's demands and needs. In this system, one of the main problems is the cold start challenge. This problem occurs when a new user logs on and because there is no sufficient information available in the system from the user, the system won't be able to provide appropriate recommendation and the system error will rise. In this paper, we propose to use a new measurement called opinion leaders to alleviate this problem. Opinion leader is a person that his opinion has an impact on the target user. As a result, in the case of a new user logging in and the user - item's matrix sparseness, we can use the opinion of opinion leaders to offer the appropriate recommendation for new users and thereby increase the accuracy of the recommender system. The results of several conducted tests showed that opinion leaders combined with recommender systems will effectively reduce the recommendation errors.