Ajoodha, RRosman, Benjamin S2017-12-192017-12-192017-11Ajoodha, R. and Rosman, B.S. 2017. Tracking influence between naive Bayes models using score-based structure learning. 2017 Pattern Recognition Association of South Africa and Robotics and Mechatronics International Conference (PRASA-RobMech), 29 November - 1 December 2017, Central University of Technology, Bloemfontein, Free State, South Africahttp://www.raillab.org/content/prasa-tracking-influence.pdfhttp://www.rgems.co.za/Downloads/Events/2017_PRASA-RobMech_Program.pdfhttp://hdl.handle.net/10204/9890Paper presented at the 2017 Pattern Recognition Association of South Africa and Robotics and Mechatronics International Conference (PRASA-RobMech), 29 November - 1 December 2017, Central University of Technology, Bloemfontein, Free State, South Africa. This is the accepted version of the paper.Current structure learning practices in Bayesian networks have been developed to learn the structure between observable variables and learning latent parameters independently. One exception establishes a variant of EM for learning the structure of Bayesian networks in the presence of incomplete data [1]. However, no method has demonstrated learning the influence structure between latent variables that describe (or are learned from) a number of observations. We present a method that learns a set of naive Bayes models (NBMs) independently given a partitioned set of observations, and then attempts to track the high-level influence structure between every NBM. The latent parameters of each model are then relearned to fine-tune the influence distribution between models for density estimation of new observations. Experimental results are provided which demonstrate the effectiveness of our non-parametric method. Applications of this method include knowledge discovery and density estimation in situations where we do not fully observe characteristics of the environment.enScore-based structure learningNaive Bayes modelsBayesian networksStructure scoresBayesian information criterionHeuristic searchGreedy hill-climbingExpectation maximisationStructure learningInfluence networksConference PresentationAjoodha, R., & Rosman, B. S. (2017). Tracking influence between naive Bayes models using score-based structure learning. IEEE. http://hdl.handle.net/10204/9890Ajoodha, R, and Benjamin S Rosman. "Tracking influence between naive Bayes models using score-based structure learning." (2017): http://hdl.handle.net/10204/9890Ajoodha R, Rosman BS, Tracking influence between naive Bayes models using score-based structure learning; IEEE; 2017. http://hdl.handle.net/10204/9890 .TY - Conference Presentation AU - Ajoodha, R AU - Rosman, Benjamin S AB - Current structure learning practices in Bayesian networks have been developed to learn the structure between observable variables and learning latent parameters independently. One exception establishes a variant of EM for learning the structure of Bayesian networks in the presence of incomplete data [1]. However, no method has demonstrated learning the influence structure between latent variables that describe (or are learned from) a number of observations. We present a method that learns a set of naive Bayes models (NBMs) independently given a partitioned set of observations, and then attempts to track the high-level influence structure between every NBM. The latent parameters of each model are then relearned to fine-tune the influence distribution between models for density estimation of new observations. Experimental results are provided which demonstrate the effectiveness of our non-parametric method. Applications of this method include knowledge discovery and density estimation in situations where we do not fully observe characteristics of the environment. DA - 2017-11 DB - ResearchSpace DP - CSIR KW - Score-based structure learning KW - Naive Bayes models KW - Bayesian networks KW - Structure scores KW - Bayesian information criterion KW - Heuristic search KW - Greedy hill-climbing KW - Expectation maximisation KW - Structure learning KW - Influence networks LK - https://researchspace.csir.co.za PY - 2017 T1 - Tracking influence between naive Bayes models using score-based structure learning TI - Tracking influence between naive Bayes models using score-based structure learning UR - http://hdl.handle.net/10204/9890 ER -