{"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/evaluation-measures-for-relevance-and","title":"Evaluation Measures for Relevance and Credibility in Ranked Lists","arxiv_id":"1708.07157","date":"2017-08-23","proceeding":null,"authors":["Lioma Christina","Simonsen Jakob Grue","Larsen Birger"],"abstract":"Recent discussions on alternative facts, fake news, and post truth politics\nhave motivated research on creating technologies that allow people not only to\naccess information, but also to assess the credibility of the information\npresented to them by information retrieval systems. Whereas technology is in\nplace for filtering information according to relevance and/or credibility, no\nsingle measure currently exists for evaluating the accuracy or precision (and\nmore generally effectiveness) of both the relevance and the credibility of\nretrieved results. One obvious way of doing so is to measure relevance and\ncredibility effectiveness separately, and then consolidate the two measures\ninto one. There at least two problems with such an approach: (I) it is not\ncertain that the same criteria are applied to the evaluation of both relevance\nand credibility (and applying different criteria introduces bias to the\nevaluation); (II) many more and richer measures exist for assessing relevance\neffectiveness than for assessing credibility effectiveness (hence risking\nfurther bias).\n  Motivated by the above, we present two novel types of evaluation measures\nthat are designed to measure the effectiveness of both relevance and\ncredibility in ranked lists of retrieval results. Experimental evaluation on a\nsmall human-annotated dataset (that we make freely available to the research\ncommunity) shows that our measures are expressive and intuitive in their\ninterpretation.","url_abs":"http://arxiv.org/abs/1708.07157v1","url_pdf":"http://arxiv.org/pdf/1708.07157v1.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":"evaluation-measures-for-relevance-and","repo_url":"https://github.com/diku-irlab/A66","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"information-retrieval","task_name":"Information Retrieval"},{"task_slug":"retrieval","task_name":"Retrieval"}],"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}