{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/simple-and-efficient-heterogeneous-graph","title":"Simple and Efficient Heterogeneous Graph Neural Network","arxiv_id":"2207.02547","date":"2022-07-06","proceeding":null,"authors":["Xiaocheng Yang","Mingyu Yan","Shirui Pan","Xiaochun Ye","Dongrui Fan"],"abstract":"Heterogeneous graph neural networks (HGNNs) have powerful capability to embed rich structural and semantic information of a heterogeneous graph into node representations. Existing HGNNs inherit many mechanisms from graph neural networks (GNNs) over homogeneous graphs, especially the attention mechanism and the multi-layer structure. These mechanisms bring excessive complexity, but seldom work studies whether they are really effective on heterogeneous graphs. This paper conducts an in-depth and detailed study of these mechanisms and proposes Simple and Efficient Heterogeneous Graph Neural Network (SeHGNN). To easily capture structural information, SeHGNN pre-computes the neighbor aggregation using a light-weight mean aggregator, which reduces complexity by removing overused neighbor attention and avoiding repeated neighbor aggregation in every training epoch. To better utilize semantic information, SeHGNN adopts the single-layer structure with long metapaths to extend the receptive field, as well as a transformer-based semantic fusion module to fuse features from different metapaths. As a result, SeHGNN exhibits the characteristics of simple network structure, high prediction accuracy, and fast training speed. Extensive experiments on five real-world heterogeneous graphs demonstrate the superiority of SeHGNN over the state-of-the-arts on both accuracy and training speed.","url_abs":"https://arxiv.org/abs/2207.02547v3","url_pdf":"https://arxiv.org/pdf/2207.02547v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"simple-and-efficient-heterogeneous-graph","repo_url":"https://github.com/ict-gimlab/sehgnn","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"simple-and-efficient-heterogeneous-graph","repo_url":"https://github.com/ICT-GIMLab/SeHGNN/tree/master/large","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"graph-neural-network","task_name":"Graph Neural Network"},{"task_slug":"heterogeneous-node-classification","task_name":"Heterogeneous Node Classification"},{"task_slug":"node-property-prediction","task_name":"Node Property Prediction"}],"methods":[{"method_slug":"graph-neural-network","method_name":"Graph Neural Network"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/heterogeneous-node-classification-on-acm","task":"Heterogeneous Node Classification","dataset":"ACM (Heterogeneous Node Classification)","model":"SeHGNN","rank_in_archive_order":10,"of":11,"metrics":{"Claim Classification Macro-F1":"93.95","Micro-F1":"93.87"},"uses_additional_data":false},{"leaderboard":"/sota/heterogeneous-node-classification-on-dblp-2","task":"Heterogeneous Node Classification","dataset":"DBLP (Heterogeneous Node Classification)","model":"SeHGNN","rank_in_archive_order":10,"of":11,"metrics":{"Macro-F1":"94.86","Micro-F1":"95.24"},"uses_additional_data":false},{"leaderboard":"/sota/heterogeneous-node-classification-on-freebase","task":"Heterogeneous Node Classification","dataset":"Freebase (Heterogeneous Node Classification)","model":"SeHGNN","rank_in_archive_order":8,"of":9,"metrics":{"Macro-F1":"50.71","Micro-F1":"63.41"},"uses_additional_data":false},{"leaderboard":"/sota/heterogeneous-node-classification-on-imdb","task":"Heterogeneous Node Classification","dataset":"IMDB (Heterogeneous Node Classification)","model":"SeHGNN","rank_in_archive_order":10,"of":11,"metrics":{"Macro-F1":"66.63","Micro-F1":"68.21"},"uses_additional_data":false},{"leaderboard":"/sota/heterogeneous-node-classification-on-oag-l1","task":"Heterogeneous Node Classification","dataset":"OAG-L1-Field","model":"SeHGNN","rank_in_archive_order":3,"of":5,"metrics":{"MRR":"84.95","NDCG":"86.01"},"uses_additional_data":false},{"leaderboard":"/sota/heterogeneous-node-classification-on-oag","task":"Heterogeneous Node Classification","dataset":"OAG-Venue","model":"SeHGNN","rank_in_archive_order":5,"of":5,"metrics":{"MRR":"29.11","NDCG":"46.75"},"uses_additional_data":false},{"leaderboard":"/sota/node-property-prediction-on-ogbn-mag","task":"Node Property Prediction","dataset":"ogbn-mag","model":"SeHGNN (ComplEx embs)","rank_in_archive_order":10,"of":39,"metrics":{"Ext. data":"No","Number of params":"8371231","Test Accuracy":"0.5719 ± 0.0012","Validation Accuracy":"0.5917 ± 0.0009"},"uses_additional_data":false},{"leaderboard":"/sota/node-property-prediction-on-ogbn-mag","task":"Node Property Prediction","dataset":"ogbn-mag","model":"SeHGNN","rank_in_archive_order":11,"of":39,"metrics":{"Ext. data":"No","Number of params":"8371231","Test Accuracy":"0.5671 ± 0.0014","Validation Accuracy":"0.5870 ± 0.0008"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2207.02547","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}