Abstract-Online reviews have become an increasingly important resource for decision making and product designing. But reviews systems are often targeted by opinion spamming. Although fake review detection has been studied by researchers for years using supervised learning, ground truth of large scale datasets is still unavailable and most of existing approaches of supervised learning are based on pseudo fake reviews rather than real fake reviews. Working with Dianping 1 , the largest Chinese review hosting site, we present the first reported work on fake review detection in Chinese with filtered reviews from Dianping's fake review detection system. Dianping's algorithm has a very high precision, but the recall is hard to know. This means that all fake reviews detected by the system are almost certainly fake but the remaining reviews (unknown set) may not be all genuine. Since the unknown set may contain many fake reviews, it is more appropriate to treat it as an unlabeled set. This calls for the model of learning from positive and unlabeled examples (PU learning). By leveraging the intricate dependencies among reviews, users and IP addresses, we first propose a collective classification algorithm called Multi-typed Heterogeneous Collective Classification (MHCC) and then extend it to Collective Positive and Unlabeled learning (CPU). Our experiments are conducted on real-life reviews of 500 restaurants in Shanghai, China. Results show that our proposed models can markedly improve the F1 scores of strong baselines in both PU and non-PU learning settings. Since our models only use language independent features, they can be easily generalized to other languages.
Fake review detection has been studied by researchers for several years. However, so far all reported studies are based on English reviews. This paper reports a study of detecting fake reviews in Chinese. Our review dataset is from the Chinese review hosting site Dianping 1 , which has built a fake review detection system. They are confident that their algorithm has a very high precision, but they don't know the recall. This means that all fake reviews detected by the system are almost certainly fake but the remaining reviews may not be all genuine. This paper first reports a supervised learning study of two classes, fake and unknown. However, since the unknown set may contain many fake reviews, it is more appropriate to treat it as an unlabeled set. This calls for the model of learning from positive and unlabeled examples (or PU-learning). Experimental results show that PU learning not only outperforms supervised learning significantly, but also detects a large number of potentially fake reviews hidden in the unlabeled set that Dianping fails to detect.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.