{"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/listening-between-the-lines-learning-personal","title":"Listening between the Lines: Learning Personal Attributes from Conversations","arxiv_id":"1904.10887","date":"2019-04-24","proceeding":null,"authors":["Anna Tigunova","Andrew Yates","Paramita Mirza","Gerhard Weikum"],"abstract":"Open-domain dialogue agents must be able to converse about many topics while\nincorporating knowledge about the user into the conversation. In this work we\naddress the acquisition of such knowledge, for personalization in downstream\nWeb applications, by extracting personal attributes from conversations. This\nproblem is more challenging than the established task of information extraction\nfrom scientific publications or Wikipedia articles, because dialogues often\ngive merely implicit cues about the speaker. We propose methods for inferring\npersonal attributes, such as profession, age or family status, from\nconversations using deep learning. Specifically, we propose several Hidden\nAttribute Models, which are neural networks leveraging attention mechanisms and\nembeddings. Our methods are trained on a per-predicate basis to output rankings\nof object values for a given subject-predicate combination (e.g., ranking the\ndoctor and nurse professions high when speakers talk about patients, emergency\nrooms, etc). Experiments with various conversational texts including Reddit\ndiscussions, movie scripts and a collection of crowdsourced personal dialogues\ndemonstrate the viability of our methods and their superior performance\ncompared to state-of-the-art baselines.","url_abs":"http://arxiv.org/abs/1904.10887v1","url_pdf":"http://arxiv.org/pdf/1904.10887v1.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":"listening-between-the-lines-learning-personal","repo_url":"https://github.com/Anna146/HiddenAttributeModels","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"articles","task_name":"Articles"},{"task_slug":"attribute","task_name":"Attribute"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1904.10887","atlas_url":"https://app.syntology.ai/?focus=1904.10887","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}