Predicting Trust Using Reputation Features

Recommender systems are increasingly being used in e-commerce websites for solving the problem of finding right kind of information. Collaborative filtering is considered as most promising method for recommendation because it recommends items based on common interests of users. Different elements like trust and emotion are being considered in research for improving recommendation accuracy.
In trust aware recommender systems, trusted users are used for recommending an item to an active user. From literature, it is proven that including all trusted users in recommendation process reduces its performance so this research work performs a filtration process on users for reduction of trusted neighborhood of an active user. The main idea of this research work is to keep only those users in trusted neighborhood whose rating behavior is similar to an active user. For this purpose, items are divided into three different subspaces which are interested items (if rating is 4 to 5), NIU items (if rating is 3 or 3.5) and uninterested items (if rating is in range of 0.5 to 2.5). A trusted user will only be included in recommendation process if its subspaces matches to the subspaces of active user. Use of both explicit and implicit trust to calculate trust values is also focus of this research work. The suggested technique is implemented in Matlab tool and applied on FilmTrust dataset. The proposed work is assessed by some quality measures such as Mean Absolute Error (MAE), Recommendation Coverage (RC) and User Coverage (UC). The results shows that the proposed algorithm gives better results as compared to other conventional algorithms.