A Proposed Fuzzy Recommender System based on Subjective, Objective and Virtual Greedy Users’ Viewpoint

Nowadays ever-increasing number of E-commerce sites on the Internet has brought about information overload. This has made it difficult for consumers of certain products to find information about such products in an attempt to purchase products that best satisfy them. In this paper, a new hybrid system is proposed, a new collaborative filtering framework which integrates both subjective and objective information considering price and discount of items to generate recommendations for an active user. The suggested framework, which introduces a set of algorithms (namely SAPD, ASOVPD, SA and ASOV), resolves some limitations of recommender systems such as missing values and cold-start and the items that have not yet been rated by any user can be recommended. On the other hand, the type of business studied, the group shopping and discounts site, and its two important factors (price and discount percent) have been noted as innovative aspect and strengths of this dissertation. The precision of the proposed algorithm, SAPD is slightly more than those of the existing hybrid frameworks and the proposed ASOVPD algorithm performs much better for MAE than the other approaches so that employing these systems are justifiable.