Filters








13,673 Hits in 5.8 sec

MultiBiSage: A Web-Scale Recommendation System Using Multiple Bipartite Graphs at Pinterest [article]

Saket Gurukar, Nikil Pancha, Andrew Zhai, Eric Kim, Samson Hu, Srinivasan Parthasarathy, Charles Rosenberg, Jure Leskovec
2022 arXiv   pre-print
MultiBiSage can capture the graph structure of multiple bipartite graphs to learn high-quality pin embeddings.  ...  These embeddings can then be used for several tasks such as recommendation and search.  ...  Both DistDGL and Aligraph rely on multi-processing for "on-thefly" sampling of node's neighbors.  ... 
arXiv:2205.10666v1 fatcat:3pf3hwzyc5hubfqjqifmefh2zy

When Product Search Meets Collaborative Filtering: A Hierarchical Heterogeneous Graph Neural Network Approach [article]

Xiangkun Yin, Yangyang Guo, Liqiang Nie, Zhiyong Cheng
2021 arXiv   pre-print
To close the gap between collaborative filtering and product search, we propose a Hierarchical Heterogeneous Graph Neural Network (HHGNN) approach in this paper.  ...  The sequence edge accounts for the syntax formulation from word nodes to sentence nodes; the composition edge aggregates the semantic features to the user and product nodes; and the interaction edge on  ...  The spectral-based methods define the convolution operation on the basis of the graph signal processing theory [26] . For instance, Bruna et al.  ... 
arXiv:2108.07574v2 fatcat:k2fxc3bonvaxlmo2em46w5es3m

Efficient and Effective Multi-Modal Queries through Heterogeneous Network Embedding

Chi Thang Duong, Tam Thanh Nguyen, Hongzhi Yin, Matthias Weidlich, Son Mai, Karl Aberer, Quoc Viet Hung Nguyen
2021 IEEE Transactions on Knowledge and Data Engineering  
CONCLUSIONS In this paper, we presented a new direction for multimodal IR that relies on an embedding of a heterogeneous information network.  ...  Based thereon, we proposed a novel network embedding model to obtain a vectorized representation of a HIN and a technique to construct a query embedding based on the HIN embedding through mapping and combination  ...  Two nodes are considered to be close, if one occurs on the random walk from the other one. Network embedding based on random walks can be considered as an application of word embeddings for graphs.  ... 
doi:10.1109/tkde.2021.3052871 fatcat:eff3dykxhbhgzeywvpfkjswcju

Multi-hop Reading Comprehension across Multiple Documents by Reasoning over Heterogeneous Graphs [article]

Ming Tu, Guangtao Wang, Jing Huang, Yun Tang, Xiaodong He, Bowen Zhou
2019 arXiv   pre-print
We employ Graph Neural Networks (GNN) based message passing algorithms to accumulate evidences on the proposed HDE graph.  ...  We introduce a heterogeneous graph with different types of nodes and edges, which is named as Heterogeneous Document-Entity (HDE) graph.  ...  Acknowledgements We would like to thank Johannes Welbl for running evaluation on our submitted model.  ... 
arXiv:1905.07374v2 fatcat:epznpgcp7vgovminxdcn53t5yq

Uniting Heterogeneity, Inductiveness, and Efficiency for Graph Representation Learning

Tong Chen, Hongzhi Yin, Jie Ren, Zi Huang, Xiangliang Zhang, Hao Wang
2021 IEEE Transactions on Knowledge and Data Engineering  
Recently, bearing the message passing paradigm, graph neural networks (GNNs) have greatly advanced the performance of node representation learning on graphs.  ...  However, a majority class of GNNs are only designed for homogeneous graphs, leading to inferior adaptivity to the more informative heterogeneous graphs with various types of nodes and edges.  ...  In this paper, we focus on semisupervised node classification, which is one of the most popular applications for heterogeneous graph embedding models [5] - [7] .  ... 
doi:10.1109/tkde.2021.3100529 fatcat:azur4cdwafg3zhk27tnssqysza

Heterogeneous Graph Attention Network for Multi-hop Machine Reading Comprehension [article]

Feng Gao, Jian-Cheng Ni, Peng Gao, Zi-Li Zhou, Yan-Yan Li, Hamido Fujita
2021 arXiv   pre-print
The name "crname" is a metaphor for the pattern of the model: regard the subjects of queries as the start points of clues, take the reasoning entities as bridge points, and consider the latent candidate  ...  Spectral models based on graph convolutional networks grant the inferring abilities and lead to competitive results, however, part of them still face the challenge of analyzing the reasoning in a human-understandable  ...  ACKNOWLEDGMENT The authors would like to thank the UCL machine reading group who creates the QANGAROO dataset and their help for evaluating our model.  ... 
arXiv:2107.00841v1 fatcat:d72l6nda5vgpbn2pz7djddfv4u

Graph Neural Networks in Real-Time Fraud Detection with Lambda Architecture [article]

Mingxuan Lu, Zhichao Han, Zitao Zhang, Yang Zhao, Yinan Shan
2021 arXiv   pre-print
Dynamic Snapshot (DDS) linkage design for graph construction and a Lambda Neural Networks (LNN) architecture for effective inference with Graph Neural Networks embeddings.  ...  In order to leverage graph networks to decrease fraud rate efficiently and guarantee the information flow passed through neighbors only from the past of the checkouts, we first present a novel Directed  ...  Conclusions We present an approach LNN on DDS graph, to leverage both graph structure and time snapshot features for fraud transaction detection.  ... 
arXiv:2110.04559v1 fatcat:pcvlklnhnnesvek4rnbanekxhq

Uniting Heterogeneity, Inductiveness, and Efficiency for Graph Representation Learning [article]

Tong Chen, Hongzhi Yin, Jie Ren, Zi Huang, Xiangliang Zhang, Hao Wang
2021 arXiv   pre-print
Recently, bearing the message passing paradigm, graph neural networks (GNNs) have greatly advanced the performance of node representation learning on graphs.  ...  However, a majority class of GNNs are only designed for homogeneous graphs, leading to inferior adaptivity to the more informative heterogeneous graphs with various types of nodes and edges.  ...  In this paper, we focus on semisupervised node classification, which is one of the most popular applications for heterogeneous graph embedding models [5] - [7] .  ... 
arXiv:2104.01711v2 fatcat:uozthcnesjbszdf2ciauvvgrai

GSim: A Graph Neural Network based Relevance Measure for Heterogeneous Graphs [article]

Linhao Luo, Yixiang Fang, Moli Lu, Xin Cao, Xiaofeng Zhang, Wenjie Zhang
2022 arXiv   pre-print
Most of existing relevance measures focus on homogeneous networks where objects are of the same type, and a few measures are developed for heterogeneous graphs, but they often need the pre-defined meta-path  ...  Defining meaningful meta-paths requires much domain knowledge, which largely limits their applications, especially on schema-rich heterogeneous graphs like knowledge graphs.  ...  For instance, ESim [11] learns node embeddings under the guidance of meta-paths and uses the embeddings for relevance measure in heterogeneous graphs.  ... 
arXiv:2208.06144v1 fatcat:whp7xd6w2rgfvhpoe4n53prdju

Multi-hop Reading Comprehension across Multiple Documents by Reasoning over Heterogeneous Graphs

Ming Tu, Guangtao Wang, Jing Huang, Yun Tang, Xiaodong He, Bowen Zhou
2019 Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics  
We employ Graph Neural Networks (GNN) based message passing algorithms to accumulate evidences on the proposed HDE graph.  ...  We introduce a heterogeneous graph with different types of nodes and edges, which is named as Heterogeneous Document-Entity (HDE) graph.  ...  Acknowledgements We would like to thank Johannes Welbl from University College London for running evaluation on our submitted model.  ... 
doi:10.18653/v1/p19-1260 dblp:conf/acl/TuWHTHZ19 fatcat:os2vkhkzynh5bl2xk44yzhdpkq

BRIGHT – Graph Neural Networks in Real-Time Fraud Detection [article]

Mingxuan Lu, Zhichao Han, Susie Xi Rao, Zitao Zhang, Yang Zhao, Yinan Shan, Ramesh Raghunathan, Ce Zhang, Jiawei Jiang
2022 arXiv   pre-print
Second, the latency of graph query and GNN model inference is usually up to hundreds of milliseconds, which is costly for some critical online services.  ...  For the inference stage, our speedup is on average 7.8× compared to the traditional GNN.  ...  We present a novel BRIGHT framework for efficient end-to-end learning and real-time inference without the leakage of future information.  ... 
arXiv:2205.13084v1 fatcat:vwgov5r255d2no6hvnm4zirdn4

Heterogeneous Relational Reasoning in Knowledge Graphs with Reinforcement Learning [article]

Mandana Saebi, Steven Krieg, Chuxu Zhang, Meng Jiang, Nitesh Chawla
2020 arXiv   pre-print
Experiments on real-world dataset show that our method outperforms state-of-the-art RL methods and discovers more novel paths during the training procedure.  ...  Our solution uses graph neural network (GNN) for encoding the neighborhood information and utilizes entity types to prune the action space.  ...  In light of recent work on heterogeneous networks that have demonstrated the importance of heterogeneous information [5, 36, 24, 11] and local neighborhood information [31, 37] in graph mining, we  ... 
arXiv:2003.06050v1 fatcat:jfw2bdpgrbe7lbxjgtthkg5ggi

Neural PathSim for Inductive Similarity Search in Heterogeneous Information Networks [article]

Wenyi Xiao, Huan Zhao, Vincent W. Zheng, Yangqiu Song
2021 arXiv   pre-print
PathSim is a widely used meta-path-based similarity in heterogeneous information networks. Numerous applications rely on the computation of PathSim, including similarity search and clustering.  ...  Computing PathSim scores on large graphs is computationally challenging due to its high time and storage complexity.  ...  ACKNOWLEDGEMENTS The authors of this paper were supported the NSFC Fund (U20B2053 ) from the NSFC of China, the RIF (R6020-19 and R6021-20) and the GRF (16211520) from RGC of Hong Kong, the MHKJFS (MHP  ... 
arXiv:2109.01549v1 fatcat:tnhpw3onebdwdlgf4waibtwa44

Context-aware Heterogeneous Graph Attention Network for User Behavior Prediction in Local Consumer Service Platform [article]

Peiyuan Zhu, Xiaofeng Wang, Zisen Sang, Aiquan Yuan, Guodong Cao
2021 arXiv   pre-print
Hence, in this paper, we propose a context-aware heterogeneous graph attention network (CHGAT) to dynamically generate the representation of the user and to estimate the probability for future behavior  ...  Specifically, we first construct the meta-path based heterogeneous graphs with the historical behaviors from multiple sources and comprehend heterogeneous vertices in the graph with a novel unified knowledge  ...  And the detailed aggregation process of our constructed heterogeneous graph is then formulated.  ... 
arXiv:2106.14652v2 fatcat:wlsiou5cyvbsbdlavhlxmxeodq

A Survey on Heterogeneous Graph Embedding: Methods, Techniques, Applications and Sources [article]

Xiao Wang and Deyu Bo and Chuan Shi and Shaohua Fan and Yanfang Ye and Philip S. Yu
2020 arXiv   pre-print
survey and categorize the state-of-the-art HG embedding methods based on the information they used in the learning process to address the challenges posed by the HG heterogeneity.  ...  We first introduce the basic concepts of HG and discuss the unique challenges brought by the heterogeneity for HG embedding in comparison with homogeneous graph representation learning; and then we systemically  ...  It constructs a useritem-query heterogeneous graph and designs a meta-pathguided HGNN to learn the embedding of users, items and queries, which can capture the intent of users.  ... 
arXiv:2011.14867v1 fatcat:phfoxj7qsrfshfednomeok7pau
« Previous Showing results 1 — 15 out of 13,673 results