Most of the present day recommender systems recommend the products to the buyers based on their personalized preferences and product information. However, these systems lack the ability to recognize and incorporate buyer’s cognitive and emotional characteristics. With an aim to address these limitations, the present study proposes a new methodology for online product recommendation based on the concepts of fuzzy sets. The proposed methodology takes into account multiple attributes of the products which are non-commensurable, conflicting and fuzzy in nature and recommends the most suitable product to the buyer on the basis of her/his desire on attribute satisfactions. The novelty of the methodology lies in its ability to derive and assimilate buyer’s cognitive characteristics such as tranquillity (anxiety), and attitudinal flexibility in the recommendation process as they are vital in choosing the right product as per buyer desire. The methodology specifically uses Maximal Entropy Ordered Weighted Averaging (MEOWA) operator to gather maximum information about the buyer while objectively deriving the parametric value β. Additionally, the methodology recognizes the buyers’ relative product preferences through a reorder vector which is subsequently used as indices for attribute aggregation using Induced Ordered Weighted Averaging (IOWA) operator. An algorithm Prod_Ranking() is written in our paper and implemented in “Python” to verify the robustness of our procedure. A numerical example of a car-purchasing problem is also illustrated to highlight the procedure developed. At the end, the proposed methodology is compared with other similar works and its advantages are well established.