{"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/semre-rank-improving-automatic-term","title":"SemRe-Rank: Improving Automatic Term Extraction By Incorporating Semantic Relatedness With Personalised PageRank","arxiv_id":"1711.03373","date":"2017-11-09","proceeding":null,"authors":["Ziqi Zhang","Jie Gao","Fabio Ciravegna"],"abstract":"Automatic Term Extraction deals with the extraction of terminology from a\ndomain specific corpus, and has long been an established research area in data\nand knowledge acquisition. ATE remains a challenging task as it is known that\nthere is no existing ATE methods that can consistently outperform others in any\ndomain. This work adopts a refreshed perspective to this problem: instead of\nsearching for such a 'one-size-fit-all' solution that may never exist, we\npropose to develop generic methods to 'enhance' existing ATE methods. We\nintroduce SemRe-Rank, the first method based on this principle, to incorporate\nsemantic relatedness - an often overlooked venue - into an existing ATE method\nto further improve its performance. SemRe-Rank incorporates word embeddings\ninto a personalised PageRank process to compute 'semantic importance' scores\nfor candidate terms from a graph of semantically related words (nodes), which\nare then used to revise the scores of candidate terms computed by a base ATE\nalgorithm. Extensively evaluated with 13 state-of-the-art base ATE methods on\nfour datasets of diverse nature, it is shown to have achieved widespread\nimprovement over all base methods and across all datasets, with up to 15\npercentage points when measured by the Precision in the top ranked K candidate\nterms (the average for a set of K's), or up to 28 percentage points in F1\nmeasured at a K that equals to the expected real terms in the candidates (F1 in\nshort). Compared to an alternative approach built on the well-known TextRank\nalgorithm, SemRe-Rank can potentially outperform by up to 8 points in Precision\nat top K, or up to 17 points in F1.","url_abs":"http://arxiv.org/abs/1711.03373v3","url_pdf":"http://arxiv.org/pdf/1711.03373v3.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":"semre-rank-improving-automatic-term","repo_url":"https://github.com/ziqizhang/data","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"semre-rank-improving-automatic-term","repo_url":"https://github.com/ziqizhang/semrerank","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"semre-rank-improving-automatic-term","repo_url":"https://github.com/kevinlu1248/pyate","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"term-extraction","task_name":"Term Extraction"},{"task_slug":"word-embeddings","task_name":"Word Embeddings"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}