Optimizing hidden Markov models with a genetic algorithm

A review and experimental study on the application of classifiers and evolutionary algorithms in EEG-based brain–machine interface systems

“…Reference [202] uses a hidden Markov model (HMM) classifier based on the HS algorithm for classification of motor imagery ERP signals. Success or failure of the HMM algorithm largely depends on the type of structure and parameters used in the algorithm [203], therefore, HS is used to train HMM. Usually, not using optimal parameters causes over-fitting in HMM [202].…”

A novel two-level particle swarm optimization approach to train the transformational grammar based hidden Markov models for performing structural alignment of pseudoknotted RNA

“…The modeling of MSA is performed by training of conventional HMMs, which is a computationally hard problem. Conventional HMMs are generally trained by Baum-Welch algorithm [45] and recently by plentiful meta-heuristic approaches such as evolutionary algorithms [46][47][48], simulating annealing [49,50] and PSO [51,52].…”

The evolutionary computation approach to motif discovery in biological sequences

“…Most of these have been concerned with multiple sequence alignments [1,12,13,17,18,25,27,32,36,37,42,44,[50][51][52][53][57][58][59]64,66,68,73], although a number have targeted other representations [20,28,29,33,34,39,40,56,60,63,65,71,72]. Most MSA approaches can be divided into two classes: those which directly evolve alignments e.g.…”

“…Thomsen [63] and Won et al [71] have used similar approaches for other biosequence applications. EAs have also been used to evolve HMMs for non-biological sequence classification problems [16,41,60]. In [16], a GA is used to optimise HMM parameter values, and in [41] a GA is used to optimise both topology and parameter values -in both cases leading to more accurate models than those produced using Baum-Welch.…”