Multi-Aspect Feature Based Neural Network Model in Detecting Fake Reviews
Ning Luo, Huaxun Deng, Linfeng Zhao, Yuan Liu, Xingwei Wang, Zhenhua Tan · 2017
With the development of network technology and e-commerce, online-purchasing has become a fashion which takes a significant ratio of the whole market. Product reviews in e-market platform have a lot of information, and buyers tend to rely on the product'information and the reviews to determine the exactly quality of the product. However, the existence of fake reviews will mislead the consumers and result in property losses and unsatisfactory transaction experience, and even the failure of the e-market platform in the long run. Therefore, it is significant to detect the fake reviews effectively. In this paper we propose a multi-aspect based neural network model to identify fake reviews, where the review features in the three aspects: metadata, similarity, and sentiment, are considered. We further establish the supervisory model and use the neural network training classifier to identify the fake reviews. Our experimental results show that this method has a great recognition rate and recall rate.