{"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/atp-directed-graph-embedding-with-asymmetric","title":"ATP: Directed Graph Embedding with Asymmetric Transitivity Preservation","arxiv_id":"1811.00839","date":"2018-11-02","proceeding":null,"authors":["Jiankai Sun","Bortik Bandyopadhyay","Armin Bashizade","Jiongqian Liang","P. Sadayappan","Srinivasan Parthasarathy"],"abstract":"Directed graphs have been widely used in Community Question Answering\nservices (CQAs) to model asymmetric relationships among different types of\nnodes in CQA graphs, e.g., question, answer, user. Asymmetric transitivity is\nan essential property of directed graphs, since it can play an important role\nin downstream graph inference and analysis. Question difficulty and user\nexpertise follow the characteristic of asymmetric transitivity. Maintaining\nsuch properties, while reducing the graph to a lower dimensional vector\nembedding space, has been the focus of much recent research. In this paper, we\ntackle the challenge of directed graph embedding with asymmetric transitivity\npreservation and then leverage the proposed embedding method to solve a\nfundamental task in CQAs: how to appropriately route and assign newly posted\nquestions to users with the suitable expertise and interest in CQAs. The\ntechnique incorporates graph hierarchy and reachability information naturally\nby relying on a non-linear transformation that operates on the core\nreachability and implicit hierarchy within such graphs. Subsequently, the\nmethodology levers a factorization-based approach to generate two embedding\nvectors for each node within the graph, to capture the asymmetric transitivity.\nExtensive experiments show that our framework consistently and significantly\noutperforms the state-of-the-art baselines on two diverse real-world tasks:\nlink prediction, and question difficulty estimation and expert finding in\nonline forums like Stack Exchange. Particularly, our framework can support\ninductive embedding learning for newly posted questions (unseen nodes during\ntraining), and therefore can properly route and assign these kinds of questions\nto experts in CQAs.","url_abs":"http://arxiv.org/abs/1811.00839v2","url_pdf":"http://arxiv.org/pdf/1811.00839v2.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":"atp-directed-graph-embedding-with-asymmetric","repo_url":"https://github.com/zhenv5/atp","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"community-question-answering","task_name":"Community Question Answering"},{"task_slug":"graph-embedding","task_name":"Graph Embedding"},{"task_slug":"link-prediction","task_name":"Link Prediction"},{"task_slug":"question-answering","task_name":"Question Answering"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/link-prediction-on-cit-hepph","task":"Link Prediction","dataset":"Cit-HepPH","model":"Asymmetric Transitivity Preservation","rank_in_archive_order":1,"of":1,"metrics":{"AUC":"89.16"},"uses_additional_data":false},{"leaderboard":"/sota/link-prediction-on-gnutella","task":"Link Prediction","dataset":"Gnutella","model":"Asymmetric Transitivity Preservation","rank_in_archive_order":1,"of":1,"metrics":{"AUC":"93.14"},"uses_additional_data":false},{"leaderboard":"/sota/link-prediction-on-wiki-vote","task":"Link Prediction","dataset":"Wiki-Vote","model":"Asymmetric Transitivity Preservation","rank_in_archive_order":1,"of":1,"metrics":{"AUC":"94.81"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}