{"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/predicting-multiple-demographic-attributes","title":"Predicting Multiple Demographic Attributes with Task Specific Embedding Transformation and Attention Network","arxiv_id":"1903.10144","date":"2019-03-25","proceeding":null,"authors":["Raehyun Kim","Hyunjae Kim","Janghyuk Lee","Jaewoo Kang"],"abstract":"Most companies utilize demographic information to develop their strategy in a\nmarket. However, such information is not available to most retail companies.\nSeveral studies have been conducted to predict the demographic attributes of\nusers from their transaction histories, but they have some limitations. First,\nthey focused on parameter sharing to predict all attributes but capturing\ntask-specific features is also important in multi-task learning. Second, they\nassumed that all transactions are equally important in predicting demographic\nattributes. However, some transactions are more useful than others for\npredicting a certain attribute. Furthermore, decision making process of models\ncannot be interpreted as they work in a black-box manner. To address the\nlimitations, we propose an Embedding Transformation Network with Attention\n(ETNA) model which shares representations at the bottom of the model structure\nand transforms them to task-specific representations using a simple linear\ntransformation method. In addition, we can obtain more informative transactions\nfor predicting certain attributes using the attention mechanism. The\nexperimental results show that our model outperforms the previous models on all\ntasks. In our qualitative analysis, we show the visualization of attention\nweights, which provides business managers with some useful insights.","url_abs":"http://arxiv.org/abs/1903.10144v1","url_pdf":"http://arxiv.org/pdf/1903.10144v1.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":"predicting-multiple-demographic-attributes","repo_url":"https://github.com/dmis-lab/demographic-prediction","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"attribute","task_name":"Attribute"},{"task_slug":"decision-making","task_name":"Decision Making"},{"task_slug":"multi-task-learning","task_name":"Multi-Task 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}