International Journal of Recent Trends in Engineering & Research

online ISSN


Publication Date : 31/03/2016

Author(s) :

S.P.VIDHYAPRIYA , Safia mahmoodha. S , Saranya.P , Boopathi.A , Sabarisan.G.

Volume/Issue :
Volume 2
Issue 3
(03 - 2016)

Abstract :

E-Commerce presents a greater opportunity for people to purchase products online at cheaper price. In this E-Commerce service there exists few problems in return of payment and breakage issues. It also provides a platform for users to share their consumption experience and quality of product. However, at the same time we face the information overloading problem and existence of fake contents. How to filter trusted information from these reviews and make an accurate recommendation and to solve issues in services is difficult for us. Here, we firstly propose analyze scheme which eliminates fake contents and it calculates each user’s score based on real reviews. In addition to this, we consider rating scores from user based on the specifications of product and its performance. Secondly, a platform for users which allows users to purchase based on their geographical location. At last, we combine both factors into our recommender system to make an accurate rating prediction, which is based on probabilistic matrix factorization. We conduct a series of experiments that shows accurate results and needed platform for both customers and

No. of Downloads :