{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,12]],"date-time":"2026-03-12T14:29:11Z","timestamp":1773325751546,"version":"3.50.1"},"reference-count":19,"publisher":"World Scientific Pub Co Pte Lt","issue":"05","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Int. J. Artif. Intell. Tools"],"published-print":{"date-parts":[[2009,10]]},"abstract":"<jats:p> Maximum likelihood method for estimating parameters of Bayesian networks (BNs) is efficient and accurate for large samples. However, the method suffers from overfitting when the sample size is small. Bayesian methods, which are effective to avoid overfitting, present difficulties for determining optimal hyperparameters of prior distributions with good balance between theoretical and practical points of view when no prior knowledge is available. <\/jats:p><jats:p> As described in this paper, we propose an alternative estimation method of the parameters on BNs. The method uses a principle, rooted in thermodynamics, of minimizing free energy (MFE). We define internal energies, entropies, and temperature, which constitute free energies. Especially for temperature, we propose a \"data temperature\" assumption and some explicit models. This approach can treat the maximum likelihood principle and the maximum entropy principle in a unified manner of the MFE principle. For assessments of classification accuracy, our method shows higher accuracy than that obtained using the Bayesian method with normally recommended hyperparameters. Moreover, our method exhibits robustness for the choice of introduced hyperparameters. <\/jats:p>","DOI":"10.1142\/s0218213009000342","type":"journal-article","created":{"date-parts":[[2009,10,30]],"date-time":"2009-10-30T09:01:16Z","timestamp":1256893276000},"page":"653-671","source":"Crossref","is-referenced-by-count":11,"title":["\"DATA TEMPERATURE\" IN MINIMUM FREE ENERGIES FOR PARAMETER LEARNING OF BAYESIAN NETWORKS"],"prefix":"10.1142","volume":"18","author":[{"given":"TAKASHI","family":"ISOZAKI","sequence":"first","affiliation":[{"name":"Graduate School of Information Systems, The University of Electro-Communications, 1-5-1 Chofugaoka, Chofu-shi, Tokyo 182-8585, Japan"},{"name":"Research and Technology Group, Fuji Xerox Co. Ltd., 3-1-1 Roppongi, Minato-ku, Tokyo 106-0032, Japan"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"NORIJI","family":"KATO","sequence":"additional","affiliation":[{"name":"Research and Technology Group, Fuji Xerox Co. Ltd., 430 Sakai, Nakai-machi, Ashigarakami-gun, Kanagawa 259-0157, Japan"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"MAOMI","family":"UENO","sequence":"additional","affiliation":[{"name":"Graduate School of Information Systems, The University of Electro-Communications, 1-5-1 Chofugaoka, Chofu-shi, Tokyo 182-8585, Japan"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"219","published-online":{"date-parts":[[2011,11,21]]},"reference":[{"key":"rf1","doi-asserted-by":"publisher","DOI":"10.1007\/978-0-387-68282-2"},{"key":"rf2","volume-title":"Probabilistic Reasoning in Intelligent Systems","author":"Pearl J.","year":"1988"},{"key":"rf3","first-page":"90","volume":"31","author":"Heckerman D.","journal-title":"Methods of Information in Medicine"},{"key":"rf5","doi-asserted-by":"publisher","DOI":"10.1126\/science.1094068"},{"key":"rf6","volume-title":"Bayesian Data Analysis","author":"Gelman A.","year":"2004"},{"key":"rf7","first-page":"309","volume":"9","author":"Cooper G.","journal-title":"Machine Learning"},{"key":"rf8","doi-asserted-by":"publisher","DOI":"10.1016\/0378-3758(94)90153-8"},{"key":"rf10","doi-asserted-by":"publisher","DOI":"10.1109\/TSMCA.2002.803772"},{"key":"rf11","volume-title":"Thermal Physics","author":"Kittel C.","year":"1980"},{"key":"rf14","doi-asserted-by":"publisher","DOI":"10.1093\/biomet\/85.3.549"},{"key":"rf16","doi-asserted-by":"publisher","DOI":"10.1109\/TIT.2005.850085"},{"key":"rf19","volume-title":"Learning Bayesian Networks","author":"Neapolitan R. E.","year":"2004"},{"key":"rf20","volume-title":"Thermodynamics","author":"Tazaki H.","year":"2000"},{"key":"rf21","volume-title":"Elements of Information Theory","author":"Cover T. M.","year":"2006"},{"key":"rf22","volume-title":"Method of Information Geometry","author":"Amari S.","year":"2000"},{"key":"rf23","doi-asserted-by":"publisher","DOI":"10.1002\/net.3230200507"},{"key":"rf27","doi-asserted-by":"publisher","DOI":"10.1023\/A:1007465528199"},{"key":"rf28","volume-title":"Causation, Prediction and Search","author":"Spirtes P.","year":"2000"},{"key":"rf29","doi-asserted-by":"publisher","DOI":"10.1016\/S0004-3702(02)00191-1"}],"container-title":["International Journal on Artificial Intelligence Tools"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.worldscientific.com\/doi\/pdf\/10.1142\/S0218213009000342","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2019,8,7]],"date-time":"2019-08-07T02:28:24Z","timestamp":1565144904000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.worldscientific.com\/doi\/abs\/10.1142\/S0218213009000342"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2009,10]]},"references-count":19,"journal-issue":{"issue":"05","published-online":{"date-parts":[[2011,11,21]]},"published-print":{"date-parts":[[2009,10]]}},"alternative-id":["10.1142\/S0218213009000342"],"URL":"https:\/\/doi.org\/10.1142\/s0218213009000342","relation":{},"ISSN":["0218-2130","1793-6349"],"issn-type":[{"value":"0218-2130","type":"print"},{"value":"1793-6349","type":"electronic"}],"subject":[],"published":{"date-parts":[[2009,10]]}}}