{"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/finding-streams-in-knowledge-graphs-to","title":"Finding Streams in Knowledge Graphs to Support Fact Checking","arxiv_id":"1708.07239","date":"2017-08-24","proceeding":null,"authors":["Prashant Shiralkar","Alessandro Flammini","Filippo Menczer","Giovanni Luca Ciampaglia"],"abstract":"The volume and velocity of information that gets generated online limits\ncurrent journalistic practices to fact-check claims at the same rate.\nComputational approaches for fact checking may be the key to help mitigate the\nrisks of massive misinformation spread. Such approaches can be designed to not\nonly be scalable and effective at assessing veracity of dubious claims, but\nalso to boost a human fact checker's productivity by surfacing relevant facts\nand patterns to aid their analysis. To this end, we present a novel,\nunsupervised network-flow based approach to determine the truthfulness of a\nstatement of fact expressed in the form of a (subject, predicate, object)\ntriple. We view a knowledge graph of background information about real-world\nentities as a flow network, and knowledge as a fluid, abstract commodity. We\nshow that computational fact checking of such a triple then amounts to finding\na \"knowledge stream\" that emanates from the subject node and flows toward the\nobject node through paths connecting them. Evaluation on a range of real-world\nand hand-crafted datasets of facts related to entertainment, business, sports,\ngeography and more reveals that this network-flow model can be very effective\nin discerning true statements from false ones, outperforming existing\nalgorithms on many test cases. Moreover, the model is expressive in its ability\nto automatically discover several useful path patterns and surface relevant\nfacts that may help a human fact checker corroborate or refute a claim.","url_abs":"http://arxiv.org/abs/1708.07239v1","url_pdf":"http://arxiv.org/pdf/1708.07239v1.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":"finding-streams-in-knowledge-graphs-to","repo_url":"https://github.com/shiralkarprashant/knowledgestream","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"fact-checking","task_name":"Fact Checking"},{"task_slug":"knowledge-graphs","task_name":"Knowledge Graphs"},{"task_slug":"misinformation","task_name":"Misinformation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1708.07239","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}