No abstract

We present, GEM, the first heterogeneous graph neural network approach for detecting malicious accounts at Alipay, one of the world's leading mobile cashless payment platform. Our approach, inspired from a connected subgraph approach, adaptively learns discriminative embeddings from heterogeneous account-device graphs based on two fundamental weaknesses of attackers, i.e. device aggregation and activity aggregation. For the heterogeneous graph consists of various types of nodes, we propose an attention mechanism to learn the importance of different types of nodes, while using the sum operator for modeling the aggregation patterns of nodes in each type. Experiments show that our approaches consistently perform promising results compared with competitive methods over time.

No abstract

We present, GeniePath, a scalable approach for learning adaptive receptive fields of neural networks defined on permutation invariant graph data. In GeniePath, we propose an adaptive path layer consists of two complementary functions designed for breadth and depth exploration respectively, where the former learns the importance of different sized neighborhoods, while the latter extracts and filters signals aggregated from neighbors of different hops away. Our method works in both transductive and inductive settings, and extensive experiments compared with competitive methods show that our approaches yield state-of-the-art results on large graphs.

No abstract

The conventional single-target Cross-Domain Recommendation (CDR) only improves the recommendation accuracy on a target domain with the help of a source domain (with relatively richer information). In contrast, the novel dual-target CDR has been proposed to improve the recommendation accuracies on both domains simultaneously. However, dual-target CDR faces two new challenges: (1) how to generate more representative user and item embeddings, and (2) how to effectively optimize the user/item embeddings on each domain. To address these challenges, in this paper, we propose a graphical and attentional framework, called GA-DTCDR. In GA-DTCDR, we first construct two separate heterogeneous graphs based on the rating and content information from two domains to generate more representative user and item embeddings. Then, we propose an element-wise attention mechanism to effectively combine the embeddings of common users learned from both domains. Both steps significantly enhance the quality of user and item embeddings and thus improve the recommendation accuracy on each domain. Extensive experiments conducted on four real-world datasets demonstrate that GA-DTCDR significantly outperforms the state-of-the-art approaches.

No abstract

BACKGROUND AND PURPOSECholecystokinin (CCK) is secreted by intestinal I cells and regulates important metabolic functions. In pancreatic islets, CCK controls beta cell functions primarily through CCK 1 receptors, but the signalling pathways downstream of these receptors in pancreatic beta cells are not well defined. EXPERIMENTAL APPROACHApoptosis in pancreatic beta cell apoptosis was evaluated using Hoechst-33342 staining, TUNEL assays and Annexin-V-FITC/PI staining. Insulin secretion and second messenger production were monitored using ELISAs. Protein and phospho-protein levels were determined by Western blotting. A glucose tolerance test was carried out to examine the functions of CCK-8s in streptozotocin-induced diabetic mice. KEY RESULTSThe sulfated carboxy-terminal octapeptide CCK 26-33 amide (CCK-8s) activated CCK 1 receptors and induced accumulation of both IP 3 and cAMP. Whereas G q -PLC-IP 3 signalling was required for the CCK-8s-induced insulin secretion under low-glucose conditions, G s -PKA/Epac signalling contributed more strongly to the CCK-8s-mediated insulin secretion in high-glucose conditions. CCK-8s also promoted formation of the CCK 1 receptor/β-arrestin-1 complex in pancreatic beta cells. Using β-arrestin-1 knockout mice, we demonstrated that β-arrestin-1 is a key mediator of both CCK-8s-mediated insulin secretion and of its the protective effect against apoptosis in pancreatic beta cells. The anti-apoptotic effects of β-arrestin-1 occurred through cytoplasmic latephase ERK activation, which activates the 90-kDa ribosomal S6 kinase-phospho-Bcl-2-family protein pathway. CONCLUSIONS AND IMPLICATIONSKnowledge of different CCK 1 receptor-activated downstream signalling pathways in the regulation of distinct functions of pancreatic beta cells could be used to identify biased CCK 1 receptor ligands for the development of new anti-diabetic drugs.