{"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/few-shot-text-classification-with-pre-trained","title":"Few-Shot Text Classification with Pre-Trained Word Embeddings and a Human in the Loop","arxiv_id":"1804.02063","date":"2018-04-05","proceeding":null,"authors":["Katherine Bailey","Sunny Chopra"],"abstract":"Most of the literature around text classification treats it as a supervised\nlearning problem: given a corpus of labeled documents, train a classifier such\nthat it can accurately predict the classes of unseen documents. In industry,\nhowever, it is not uncommon for a business to have entire corpora of documents\nwhere few or none have been classified, or where existing classifications have\nbecome meaningless. With web content, for example, poor taxonomy management can\nresult in labels being applied indiscriminately, making filtering by these\nlabels unhelpful. Our work aims to make it possible to classify an entire\ncorpus of unlabeled documents using a human-in-the-loop approach, where the\ncontent owner manually classifies just one or two documents per category and\nthe rest can be automatically classified. This \"few-shot\" learning approach\nrequires rich representations of the documents such that those that have been\nmanually labeled can be treated as prototypes, and automatic classification of\nthe rest is a simple case of measuring the distance to prototypes. This\napproach uses pre-trained word embeddings, where documents are represented\nusing a simple weighted average of constituent word embeddings. We have tested\nthe accuracy of the approach on existing labeled datasets and provide the\nresults here. We have also made code available for reproducing the results we\ngot on the 20 Newsgroups dataset.","url_abs":"http://arxiv.org/abs/1804.02063v1","url_pdf":"http://arxiv.org/pdf/1804.02063v1.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":"few-shot-text-classification-with-pre-trained","repo_url":"https://github.com/katbailey/few-shot-text-classification","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"few-shot-learning","task_name":"Few-Shot Learning"},{"task_slug":"few-shot-text-classification","task_name":"Few-Shot Text Classification"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"management","task_name":"Management"},{"task_slug":"text-classification","task_name":"Text Classification"},{"task_slug":"word-embeddings","task_name":"Word Embeddings"},{"task_slug":"text-classification-1","task_name":"text-classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1804.02063","atlas_url":"https://app.syntology.ai/?focus=1804.02063","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}