{"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/accurate-parameter-estimation-for-bayesian","title":"Accurate parameter estimation for Bayesian Network Classifiers using Hierarchical Dirichlet Processes","arxiv_id":"1708.07581","date":"2017-08-25","proceeding":null,"authors":["Francois Petitjean","Wray Buntine","Geoffrey I. Webb","Nayyar Zaidi"],"abstract":"This paper introduces a novel parameter estimation method for the probability\ntables of Bayesian network classifiers (BNCs), using hierarchical Dirichlet\nprocesses (HDPs). The main result of this paper is to show that improved\nparameter estimation allows BNCs to outperform leading learning methods such as\nRandom Forest for both 0-1 loss and RMSE, albeit just on categorical datasets.\n  As data assets become larger, entering the hyped world of \"big\", efficient\naccurate classification requires three main elements: (1) classifiers with\nlow-bias that can capture the fine-detail of large datasets (2) out-of-core\nlearners that can learn from data without having to hold it all in main memory\nand (3) models that can classify new data very efficiently.\n  The latest Bayesian network classifiers (BNCs) satisfy these requirements.\nTheir bias can be controlled easily by increasing the number of parents of the\nnodes in the graph. Their structure can be learned out of core with a limited\nnumber of passes over the data. However, as the bias is made lower to\naccurately model classification tasks, so is the accuracy of their parameters'\nestimates, as each parameter is estimated from ever decreasing quantities of\ndata. In this paper, we introduce the use of Hierarchical Dirichlet Processes\nfor accurate BNC parameter estimation.\n  We conduct an extensive set of experiments on 68 standard datasets and\ndemonstrate that our resulting classifiers perform very competitively with\nRandom Forest in terms of prediction, while keeping the out-of-core capability\nand superior classification time.","url_abs":"http://arxiv.org/abs/1708.07581v3","url_pdf":"http://arxiv.org/pdf/1708.07581v3.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":"accurate-parameter-estimation-for-bayesian","repo_url":"https://github.com/fpetitjean/HierarchicalDirichletProcessEstimation","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"accurate-parameter-estimation-for-bayesian","repo_url":"https://github.com/HerrmannM/HDP","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"accurate-parameter-estimation-for-bayesian","repo_url":"https://github.com/HerrmannM/MonashMLHDP","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"accurate-parameter-estimation-for-bayesian","repo_url":"https://github.com/fpetitjean/HDP","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"parameter-estimation","task_name":"parameter estimation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}