Geilke, MichaelFrank, EibeKarwath, AndreasKramer, StefanXiong, HKarypis, GThuraisingham, BCook, DWu, X2024-12-062024-12-062013Geilke, M., Frank, E., Karwath, A., & Kramer, S. (2013). Online estimation of discrete densities. Proceedings / IEEE International Conference on Data Mining. IEEE International Conference on Data Mining, 191-200. https://doi.org/10.1109/ICDM.2013.911550-4786https://hdl.handle.net/10289/17079We address the problem of estimating a discrete joint density online, that is, the algorithm is only provided the current example and its current estimate. The proposed online estimator of discrete densities, EDDO (Estimation of Discrete Densities Online), uses classifier chains to model dependencies among features. Each classifier in the chain estimates the probability of one particular feature. Because a single chain may not provide a reliable estimate, we also consider ensembles of classifier chains and ensembles of weighted classifier chains. For all density estimators, we provide consistency proofs and propose algorithms to perform certain inference tasks. The empirical evaluation of the estimators is conducted in several experiments and on data sets of up to several million instances: We compare them to density estimates computed from Bayesian structure learners, evaluate them under the influence of noise, measure their ability to deal with concept drift, and measure the run-time performance. Our experiments demonstrate that, even though designed to work online, EDDO delivers estimators of competitive accuracy compared to batch Bayesian structure learners and batch variants of EDDO.enThis is an accepted version of a conference paper published online at https://ieeexplore.ieee.org/document/6729503. © 2013 IEEE.http://creativecommons.org/licenses/by/4.0/(ensembles of) classifier chainsBayes methodscomputer sciencedata streamsdensity estimationdensity measurementestimationHoeffding treesinference algorithmsjointsnoise measurementradiation detectorsScience & TechnologyTechnologyComputer Science, Artificial IntelligenceComputer ScienceCLASSIFICATIONConference Contribution10.1109/ICDM.2013.9146 Information and Computing Sciences4905 Statistics49 Mathematical Sciences4603 Computer Vision and Multimedia Computation4611 Machine Learning