Gaussian mixture models with equivalence constraints Academic Article uri icon

abstract

  • Abstract Gaussian Mixture Models (GMMs) have been widely used to cluster data in an unsupervised manner via the Expectation Maximization (EM) algorithm. In this chapter we suggest a semi-supervised EM algorithm that incorporates equivalence constraints into a GMM. Equivalence constraints provide information about pairs of data points, indicating whether the points arise from the same source (a must-link constraint) or from different sources (a cannot-link constraint). These constraints allow the EM algorithm to converge …

publication date

  • January 1, 2008