![Sample our Computer Science journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5928/TOC-Subject-Sample-2021-Computer-Science.jpg"})
Abstract
This study developed an adaptive electronic commerce (EC) website based on users' cognitive styles without asking users to complete any evaluation forms. In this system, a multilayer feed forward neural network (MLFF) was designed to identify the cognitive styles of anonymous users by observing their browsing behavior. Then the system presented the adaptive interfaces, designed by investigating the relationships between users' cognitive styles and browsing behavior, to users based on the identified cognitive styles. Experiments were conducted to evaluate the effectiveness of the system. The experimental results verified the potential benefits of MLFF in identifying anonymous users' cognitive styles during browsing of EC applications and provided evidence that an adaptive EC website that presents product data consistent with the users' cognitive styles can be beneficial to one-to-one Internet marketing not only for users whose cognitive styles are known before browsing but also for anonymous users whose cognitive styles are identified during browsing.
Acknowledgments
We gratefully acknowledge the research support of National Science Council of Taiwan, R.O.C. (NSC 94-2213-E-216-006; NSC 93-2213-E-216-009) and Chung-Hua University (CHU-NSC 94-2213-E-216-006; CHU-NSC 93-2213-E-216-009). We would also like to thank the anonymous reviewers for insightful comments on this article.