The primary reason for combining the mixture model and hierarchical clustering methods is that each compensates for the other's limitations.
In our comparison between the eight machine-learning models, SVM-assisted hierarchical mixture models (SV-HMM: semi-parametric, HMESVM: non-parametric) were shown to be most suitable (CCI: 67.82%, 69.38%) for the classification of HIV disease types using the glycosylation profiles (Table 2).
In order to select a reliable clustering method, the performance of six different clustering methods consisting of k-Means, Genetic k-Means, Fuzzy C-Means, Self-Organizing Map (SOM), Gaussian Mixture Model (GMM), and hierarchical model were compared.
A HME is an ensemble method for predicting the response where each model in the ensemble is weighted by probabilities estimated from a hierarchical framework of mixture models [ 18].
In this paper, we explore small-variance asymptotics for exponential family Dirichlet process (DP) and hierarchical Dirichlet process (HDP) mixture models.
In general, we can classify them into two categories: (i) simple non Gaussian models with heavy tails and (ii) mixture models with hidden variables which result to hierarchical models.
Cluster analysis methods such as k-means, hierarchical clustering, and Gaussian mixture models aim to find a partition of objects so that the objects on each subset (cluster) share some common traits.
Expectation-maximization clustering was performed using hierarchical clustering for parameterized Gaussian mixture models, setting the number of clusters to 2 (one cluster of 'identified serotonergic neurons' and one of 'unidentified neurons'), with model selection by Bayesian Information Criterion.
Then we choose the optimal model with one or two components for M i, according to the Bayesian information criterion for expectation maximization initialized by hierarchical clustering for parameterized Gaussian mixture models (Fraley and Raftery, 2002).
We extract features from this data and apply a variety of machine learning algorithms, including Gaussian mixture models and Multi-task Multi-Kernel Leareing; we are currently working to apply Bayesian hierarchical multi-task learning and Deep Learning as well.
