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.
In the actual operation of pumps, regulating the rotating speed of the pump based on the affinity law through variable speed drives is deemed as a prudent and convenient approach to mitigate energy loss. However, the multistage side channel pump is composed of one centrifugal impeller at the first stage and one or more side channel structures, the applicability of affinity law to this composite structure has not been confirmed. Three schemes with different suction angles of single-stage and one multistage side channel pump were investigated under different rotating speeds through numerical and experimental analysis. The findings elucidated that the single-stage side channel pumps exhibit a proportionate relationship to the affinity law, regardless of how the geometry varies. The numerical work was validated by the comparison between the simulated result and the tested result of the multistage side channel pump under two rotating speeds. Noticeably, the entire performance of the multistage side channel pump conforms to the affinity law, which has the same phenomenon as the single-stage side channel pump. The entropy production causing dissipation of turbulence flows in each stage exhibits a similar tendency as the overall head. As a result, the vortex distribution in average time and transient moment are almost analogous in the impeller of each stage under corresponding flow points. This briefly explains composite structures could be considered as pumps in series regardless of their composition. The outcome of this research could offer a theoretical basis for energy-saving methods of side channel pumps.
A side channel pump is a pump with a high head and a small flow that is widely used in various industrial fields. Many scientists have studied the hydraulic performance, pressure fluctuation characteristics, and gas-liquid mixed transport characteristics of this type of pump. However, these studies mainly focused on the single-stage impeller of the side channel pump, without considering the inter-stage connection channel and the multistage timing effect. These characteristics affect the hydraulic performance and pressure-pulsation characteristics of the side channel pump. Therefore, we carried out a numerical simulation and an experimental comparison on the multistage side channel pump to explore its flow characteristics during the stages. This study focused on the influence of different turbulence models on the numerical simulation of multistage side channel pumps. Shear stress transport (SST), detached eddy simulation (DES), and detached eddy simulation-curvature correction (DES-CC) turbulence models were selected for consideration. By studying the pressure and velocity streamline distribution, the turbulent kinetic energy, and the shape and volume of the vortex core area simulated by the three models, we concluded that the DES-CC model, when compared to the other models, can more fully reflect the vortex characteristics and the simulation results that are closer to the experimental data. The results of this study can be used as the basis for future research on multistage side channel pumps.
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.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.