Fuzhi Zhang

The collusive fraudulent behaviors on e-commerce platforms lead to proliferation of fraudulent reviews, which disrupt fair competition among merchants and mislead consumers’ shopping decisions. Detection of spammer groups helps purify the e-commerce environment and enhances consumers’ shopping experience. However, existing graph-based methods for detecting spammer groups first learn user node vector representations from the graph, and then use clustering methods to obtain candidate groups. Such separate two-stage detection methods are difficult to obtain high-quality candidate groups, resulting in suboptimal detection performance. Additionally, current graph construction methods used in spammer group detection do not fully consider the characteristics of spammer groups, which limits the detection performance. Aiming these concerns, we integrate heterogeneous graph attention network (HGAN) with label propagation (LP) for detecting spammer groups. First, we build a heterogeneous weighted directed (HWD) graph by analyzing the dataset and assign an initial label to each node. Then, we integrate a HGAN-module with an LP-module to obtain the HWD graph’s node embeddings and simultaneously generate candidate groups. We enhance the quality of embeddings and groups through the collaborative optimization between the predicted labels obtained from the HGAN-module and the pseudo-labels obtained from the LP-module. Finally, we calculate the suspiciousness values of groups using the reconstruction loss of the autoencoder for spammer group identification. Experiments conducted on real-world review datasets, including Amazon, Yelp, and YelpChi, demonstrate that our method achieves significant improvements in average Precision@k and Recall@k metrics compared with state-of-the-art baseline approaches.

Spammer group detection is necessary for curbing collusive review spammers on online shopping websites. However, the current detection approaches ignore exploring deep-level suspicious user review relationships and learning group features with low discrimination, which affects detection performance. Furthermore, the interpretation of detection results is easily influenced by noise features and unimportant group structures, leading to suboptimal interpretation performance. Aimed at addressing these concerns, we propose an interpretable review spammer group detection model based on knowledge distillation and counterfactual generation. First, we analyze user review information to generate a suspicious user review relationship graph, combining a graph agglomerative hierarchical clustering approach to discover candidate groups. Second, we devise a knowledge distillation network to learn discriminative candidate group features for detecting review spammer groups. Finally, we design a counterfactual generation model to search important subgraph structures for interpreting the detection results. The experiments indicate that the improvements in our model’s Precision@k and Recall@k are among the top-1000 state-of-the-art solutions on the Amazon, YelpChi, YelpNYC, and YelpZip datasets, which are [13.37%, 72.63%, 37.46%, and 18.83%] and [17.34%, 43.81%, 41.22%, and 21.05%], respectively. The Fidelities of our interpretation results under different Sparsity are around 6%, 7%, 7%, and 6% higher than that of the state-of-the-art solutions on the Amazon, YelpChi, YelpNYC, and YelpZip datasets, respectively.