{"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/merging-datasets-through-deep-learning","title":"Merging datasets through deep learning","arxiv_id":"1809.01604","date":"2018-09-05","proceeding":null,"authors":["Kavitha Srinivas","Abraham Gale","Julian Dolby"],"abstract":"Merging datasets is a key operation for data analytics. A frequent\nrequirement for merging is joining across columns that have different surface\nforms for the same entity (e.g., the name of a person might be represented as\n\"Douglas Adams\" or \"Adams, Douglas\"). Similarly, ontology alignment can require\nrecognizing distinct surface forms of the same entity, especially when\nontologies are independently developed. However, data management systems are\ncurrently limited to performing merges based on string equality, or at best\nusing string similarity. We propose an approach to performing merges based on\ndeep learning models. Our approach depends on (a) creating a deep learning\nmodel that maps surface forms of an entity into a set of vectors such that\nalternate forms for the same entity are closest in vector space, (b) indexing\nthese vectors using a nearest neighbors algorithm to find the forms that can be\npotentially joined together. To build these models, we had to adapt techniques\nfrom metric learning due to the characteristics of the data; specifically we\ndescribe novel sample selection techniques and loss functions that work for\nthis problem. To evaluate our approach, we used Wikidata as ground truth and\nbuilt models from datasets with approximately 1.1M people's names (200K\nidentities) and 130K company names (70K identities). We developed models that\nallow for joins with precision@1 of .75-.81 and recall of .74-.81. We make the\nmodels available for aligning people or companies across multiple datasets.","url_abs":"http://arxiv.org/abs/1809.01604v1","url_pdf":"http://arxiv.org/pdf/1809.01604v1.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":"merging-datasets-through-deep-learning","repo_url":"https://github.com/yehudagale/fuzzyjoiner","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"deep-learning","task_name":"Deep Learning"},{"task_slug":"management","task_name":"Management"},{"task_slug":"metric-learning","task_name":"Metric Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}