Recommender systems help the customers in choosing the right products/services by mining the opinion expressed in the user reviews. The volume of review documents available on the web is huge and it is a challenging task to read manually and summarize them. In this paper, an intuitive scoring approach is proposed for analyzing the overall opinion expressed in a review document. For the first time, an intuitive scoring (IS) approach based on the concept of hedges is used in combination with a new type of term-document matrix called as “representative term-document matrix” for summarizing the user reviews. The proposed classifier achieved a near 90% precision and accuracy of 80% for the dataset consisting of up to 538 reviews. © 2016, International Journal of Pharmacy and Technology. All rights reserved.