{"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/weisfeiler-lehman-neural-machine-for-link","title":"Weisfeiler-lehman neural machine for link prediction","arxiv_id":null,"date":"2017-08-01","proceeding":"KDD 2017 8","authors":["Muhan Zhang","Yixin Chen"],"abstract":"In this paper, we propose a next-generation link prediction method,\r\nWeisfeiler-Lehman Neural Machine (Wlnm), which learns topological features in the form of graph patterns that promote the\r\nformation of links. Wlnm has unmatched advantages including\r\nhigher performance than state-of-the-art methods and universal\r\napplicability over various kinds of networks. Wlnm extracts an\r\nenclosing subgraph of each target link and encodes the subgraph\r\nas an adjacency matrix. The key novelty of the encoding comes\r\nfrom a fast hashing-based Weisfeiler-Lehman (WL) algorithm that\r\nlabels the vertices according to their structural roles in the subgraph\r\nwhile preserving the subgraph’s intrinsic directionality. After that,\r\na neural network is trained on these adjacency matrices to learn a\r\npredictive model. Compared with traditional link prediction methods, Wlnm does not assume a particular link formation mechanism\r\n(such as common neighbors), but learns this mechanism from the\r\ngraph itself. We conduct comprehensive experiments to show that\r\nWlnm not only outperforms a great number of state-of-the-art\r\nlink prediction methods, but also consistently performs well across\r\nnetworks with different characteristics.","url_abs":"https://dl.acm.org/doi/pdf/10.1145/3097983.3097996","url_pdf":"https://dl.acm.org/doi/pdf/10.1145/3097983.3097996","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":"weisfeiler-lehman-neural-machine-for-link","repo_url":"https://github.com/muhanzhang/LinkPrediction","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"torch","reach":null}],"tasks":[{"task_slug":"link-prediction","task_name":"Link Prediction"},{"task_slug":"prediction","task_name":"Prediction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}