Many online shoppers will take a look at the reviews for the product or service they’re about to purchase. The majority will presumably trust that the e-commerce site will only be posting genuine reviews of any given product, posted by other customers. However, as several recent high-profile cases have shown this is not always the case. Unfortunately, e-commerce sites are littered with fake reviews. These can persuade innocent shoppers to make a purchase and anticipate a certain level of quality to which the product or service they receive ultimately does not reach.
Even the most respected of sites can succumb to fake reviews because it is very difficult to automate detection despite the many protections that some operators of such sites have implemented to do so. Now, writing in the International Journal of High Performance Computing and Networking, a team from China has demonstrated how a dynamic multimode network might be employed to efficiently detect fake reviews.
There are four fundamental concepts that might be examined to detect fake reviews, explain Jun Zhao and Hong Wang of the School of Information Science and Engineering, at Shandong Normal University, China. These are the quality of the merchandise, the honesty of the review, the trustworthiness of the reviewer, and the reliability of the e-commerce site. However, even taken together these cannot discern whether an unscrupulous merchant has employed third parties to post favourable but fake reviews of their products and services. In order to more subtly detect fake reviews, the team’s dynamic approach utilizes three algorithms to uncover the nuances common to fake reviews.
Zhao, J. and Wang, H. (2019) ‘Detecting fake reviews via dynamic multimode network‘, Int. J. High Performance Computing and Networking, Vol. 13, No. 4, pp.408-416.