A review of the Expectation Maximization algorithm in data-driven process identification

• Published in: Journal of process control Vol. 73; pp. 123 - 136
• Main Authors: Sammaknejad, Nima, Zhao, Yujia, Huang, Biao
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.01.2019
• Subjects: Data-driven process identification; Expectation Maximization algorithm; Hidden Markov Models; Latent variable models; Missing data; Multiple models; Outlier treatment; State space; Switching; Time delay; Data-driven process identification; Missing data; Multiple models; Latent variable models; Outlier treatment; State space; Time delay; Hidden Markov Models; Expectation Maximization algorithm; Switching

The Expectation Maximization (EM) algorithm has been widely used for parameter estimation in data-driven process identification. EM is an algorithm for maximum likelihood estimation of parameters and ensures convergence of the likelihood function. In presence of missing variables and in ill conditio...