A Content-Based Product Recommendation Approach

Abstract

In this study, a content-based recommendation approach is proposed. It utilizes the preprocessed 245 top movie summaries of IMDB and the favorite movie genres of the user elicited with the questionnaire method and then, recommends potential products -from a product pool- that the user can “like”. For testing; a test dataset that consists of real products from Amazon.com was created, and a Web application that uses the proposed approach and leads the users to evaluate the results of this approach was designed and developed. 52 volunteered subjects attended the test. The subject examined and graded each of the 10 products displayed. Grading was based on the five-level Likert-type scale “Not at all” (0%), “Slightly” (25%), “Moderate” (50%), “Very” (75%), and “Extremely” (100%). It is possible to say that the subjects are moderately liked the products. When the product evaluations are categorized in two categories as “liked” and “disliked”, it is possible to say that the subjects liked ~78.65% of the products. This approach could be integrated into e-commerce applications like Amazon.com for recommending potential products that the user can “like”.