Marketing based on user behavior on Facebook social network through recommender system design

Bahareh Shadi Shams Zamenjani


t— the influence of social networks among people and at the same time inevitable spread of commercial use of them. Accordingly, in order to sell products, recommender systems designed based on user behavior on social networks, providing a variety of commercial offers tailored to the user. The accuracy of recommender systems that make recommendations to users, and how many of the proposals are accepted by the users is important. In this paper, a recommender system is designed based on user behavior in social network Facebook in two acts and suggests that users purchase their favorite products. The first step is to examine user behavior based on user interests will be given an offer to buy products. In the second stage recommender system uses data mining techniques and suggestions to the user that is associated with their previous purchases. This is real data and the real results of it and it is valid, as well as the results show a high level of accuracy recommender system is designed to offer suggestions to users.


marketing, social network, profiles, recommender system, user modeling.

Full Text:



Maver, John, Cappy Popp; Essential Facebook Development; Addision-Wesely: page 12; 2010.

Y. A. Kim and J. Srivastava; Impact of social influence in e-commerce decision making; Technical report, ICEC 07, vol.258. ACM; New York 2007.

Jiawei Han and Micheline Kamber; Data Mining: Concepts and Techniques; Second Edition, Diane Cerra; Elsevier Inc - 13: 978-1-55860-901-3; 2006.

Lorence, Jessica; Word of Mouth Rules; Brand Strategy - pages 40 – 41; 2007.

Nong Ye (Edited) ; The Handbook Of Data Mining ; Books published by Lawrence Erlbaum Associates are printed on acid-free paper - Printed in the United States of America ; 2003.

Andrew W. Moore; Decision Trees; Professor School of Computer Science Carnegie Mellon University

B.K. Bharadwaj and S. Pal. ; Data Mining: A prediction for performance improvement using classification ; International Journal of Computer Science and Information Security (IJCSIS), Vol. 9,No.4,pp.136-140 ; 2011.