Customers make a lot of reviews on online shopping websites every day, e.g., Amazon and Taobao. Reviews affect the buying decisions of customers, meanwhile, attract lots of spammers aiming at misleading buyers. Xianyu, the largest second-hand goods app in China, suffering from spam reviews. The anti-spam system of Xianyu faces two major challenges: scalability of the data and adversarial actions taken by spammers. In this paper, we present our technical solutions to address these challenges. We propose a large-scale anti-spam method based on graph convolutional networks (GCN) for detecting spam advertisements at Xianyu, named GCN-based Anti-Spam (GAS) model. In this model, a heterogeneous graph and a homogeneous graph are integrated to capture the local context and global context of a comment. Offline experiments show that the proposed method is superior to our baseline model in which the information of reviews, features of users and items being reviewed are utilized. Furthermore, we deploy our system to process million-scale data daily at Xianyu. The online performance also demonstrates the effectiveness of the proposed method.
This study considers an original equipment manufacturer (OEM) who (a) relies on a contract manufacturer (CM) to produce its new products, and (b) engages the CM again or an independent remanufacturer (IR) to perform the remanufacturing operations of branded patented products through either authorization or outsourcing. This study chiefly investigates the optimal remanufacturing engagement strategy (i.e., engage a CM or an IR) and remanufacturing mode (i.e., authorization and outsourcing) that should be taken by an OEM. This paper derives the optimal responses of the three parties (i.e., OEM, CM, and IR) by comparing the outcomes of different remanufacturing engagement strategies and remanufacturing modes under two settings, depending on whether the wholesale price is exogenously or endogenously given. The results suggest that the OEM is better off adopting the authorization remanufacturing mode by either working with the CM or the IR when the wholesale price is exogenously given. When the wholesale price is endogenously determined by the CM, the best strategy for the OEM is to work with the IR using the outsourcing approach. The CM and the IR also show different preferences between the two modes. The IR or CM that engages in remanufacturing always prefers the outsourcing mode than the authorization mode in the two settings. However, when the CM does not engage in remanufacturing, it is better off adopting the authorization mode when the wholesale price is exogenously given or when the remanufacturing cost is relatively low in the endogenized wholesale price setting.
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.