An Accelerated Algorithm for Density Estimation in Large Databases, Using Gaussian Mixtures

An Accelerated Algorithm for Density Estimation in Large Databases, Using Gaussian Mixtures. Soto, A., Zavala, F., & Araneda, A. Cybernetics and Systems: An International Journal, 38(2):123-139, 2007.

Paper abstract bibtex 2 downloads

Today, with the advances of computer storage and technology, there are huge datasets available, offering an opportunity to extract valuable information. Probabilistic approaches are specially suited to learn from data by representing knowledge as density functions. In this paper, we choose Gaussian Mixture Models (GMMs) to represent densities, as they possess great flexibility to adequate to a wide class of problems. The classical estimation approach for GMMs corresponds to the iterative algorithm of Expectation Maximization. This approach, however, does not scale properly to meet the high demanding processing requirements of large databases. In this paper we introduce an EM-based algorithm, that solves the scalability problem. Our approach is based on the concept of data condensation which, in addition to substantially diminishing the computational load, provides sound starting values that allow the algorithm to reach convergence faster. We also focus on the model selection problem. We test our algorithm using synthetic and real databases, and find several advantages, when compared to other standard existing procedures.

@Article{	  soto:zavala:araneda:2007,
  author	= {A. Soto and F. Zavala and A. Araneda},
  title		= {An Accelerated Algorithm for Density Estimation in Large
		  Databases, Using Gaussian Mixtures},
  journal	= {Cybernetics and Systems: An International Journal},
  volume	= {38},
  number	= {2},
  pages		= {123-139},
  year		= {2007},
  abstract	= {Today, with the advances of computer storage and
		  technology, there are huge datasets available, offering an
		  opportunity to extract valuable information. Probabilistic
		  approaches are specially suited to learn from data by
		  representing knowledge as density functions. In this paper,
		  we choose Gaussian Mixture Models (GMMs) to represent
		  densities, as they possess great flexibility to adequate to
		  a wide class of problems. The classical estimation approach
		  for GMMs corresponds to the iterative algorithm of
		  Expectation Maximization. This approach, however, does not
		  scale properly to meet the high demanding processing
		  requirements of large databases. In this paper we introduce
		  an EM-based algorithm, that solves the scalability problem.
		  Our approach is based on the concept of data condensation
		  which, in addition to substantially diminishing the
		  computational load, provides sound starting values that
		  allow the algorithm to reach convergence faster. We also
		  focus on the model selection problem. We test our algorithm
		  using synthetic and real databases, and find several
		  advantages, when compared to other standard existing
		  procedures.},
  url		= {http://saturno.ing.puc.cl/media/papers_alvaro/Felipe-07.pdf}
}

Downloads: 2