Van der Walt, Christiaan MBarnard, E2012-02-142012-02-142009-11Van der Walt, C and Barnard, E. Density estimation from local structure. 20th Annual Symposium of the Pattern Recognition Association of South Africa (PRASA), Stellenbosch, South Africa, 30 November-01 December 2009, pp 131-136978-0-7992-2356-9http://www.dip.ee.uct.ac.za/prasa/PRASA2010/proceedings/2009/prasa09-23.pdfhttp://www.dip.ee.uct.ac.za/prasa/PRASA2010/proceedings/2009/http://hdl.handle.net/10204/556820th Annual Symposium of the Pattern Recognition Association of South Africa (PRASA), Stellenbosch, South Africa, 30 November-01 December 2009The authors propose a hyper-ellipsoid clustering algorithm that grows clusters from local structures in a dataset and estimates the underlying geometrical structure of data with a set of hyper-ellipsoids. The clusters are used to estimate a Gaussian Mixture Model (GMM) density function of the data and the log-likelihood scores are compared to the scores of a GMM trained with the expectation maximization (EM) algorithm on 5 real-world classification datasets (from the UCI collection). They show that their approach gives better generalization performance on unseen test sets for 4 of the 5 datasets considered.enDensity estimationHyper-ellipsoidsGaussian mixture model (GMM)Local structureConference PresentationVan der Walt, C. M., & Barnard, E. (2009). Density estimation from local structure. PRASA. http://hdl.handle.net/10204/5568Van der Walt, Christiaan M, and E Barnard. "Density estimation from local structure." (2009): http://hdl.handle.net/10204/5568Van der Walt CM, Barnard E, Density estimation from local structure; PRASA; 2009. http://hdl.handle.net/10204/5568 .TY - Conference Presentation AU - Van der Walt, Christiaan M AU - Barnard, E AB - The authors propose a hyper-ellipsoid clustering algorithm that grows clusters from local structures in a dataset and estimates the underlying geometrical structure of data with a set of hyper-ellipsoids. The clusters are used to estimate a Gaussian Mixture Model (GMM) density function of the data and the log-likelihood scores are compared to the scores of a GMM trained with the expectation maximization (EM) algorithm on 5 real-world classification datasets (from the UCI collection). They show that their approach gives better generalization performance on unseen test sets for 4 of the 5 datasets considered. DA - 2009-11 DB - ResearchSpace DP - CSIR KW - Density estimation KW - Hyper-ellipsoids KW - Gaussian mixture model (GMM) KW - Local structure LK - https://researchspace.csir.co.za PY - 2009 SM - 978-0-7992-2356-9 T1 - Density estimation from local structure TI - Density estimation from local structure UR - http://hdl.handle.net/10204/5568 ER -