{"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/link-prediction-in-networks-with-core-fringe","title":"Link Prediction in Networks with Core-Fringe Data","arxiv_id":"1811.11540","date":"2018-11-28","proceeding":null,"authors":["Austin R. Benson","Jon Kleinberg"],"abstract":"Data collection often involves the partial measurement of a larger system. A\ncommon example arises in collecting network data: we often obtain network\ndatasets by recording all of the interactions among a small set of core nodes,\nso that we end up with a measurement of the network consisting of these core\nnodes along with a potentially much larger set of fringe nodes that have links\nto the core. Given the ubiquity of this process for assembling network data, it\nis crucial to understand the role of such a `core-fringe' structure.\n  Here we study how the inclusion of fringe nodes affects the standard task of\nnetwork link prediction. One might initially think the inclusion of any\nadditional data is useful, and hence that it should be beneficial to include\nall fringe nodes that are available. However, we find that this is not true; in\nfact, there is substantial variability in the value of the fringe nodes for\nprediction. Once an algorithm is selected, in some datasets, including any\nadditional data from the fringe can actually hurt prediction performance; in\nother datasets, including some amount of fringe information is useful before\nprediction performance saturates or even declines; and in further cases,\nincluding the entire fringe leads to the best performance. While such variety\nmight seem surprising, we show that these behaviors are exhibited by simple\nrandom graph models.","url_abs":"http://arxiv.org/abs/1811.11540v2","url_pdf":"http://arxiv.org/pdf/1811.11540v2.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":"link-prediction-in-networks-with-core-fringe","repo_url":"https://github.com/arbenson/cflp","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"link-prediction","task_name":"Link Prediction"},{"task_slug":"prediction","task_name":"Prediction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1811.11540","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}