Improved initialisation of model-based clustering using Gaussian hierarchical partitions

Initialisation of the EM algorithm in model-based clustering is often crucial. Various starting points in the parameter space often lead to different local maxima of the likelihood function and, so to different clustering partitions. Among the several approaches available in the literature, model-based agglomerative hierarchical clustering is used to provide initial partitions in the popular mclustR package. This choice is computationally convenient and often yields good clustering partitions. However, in certain circumstances, poor initial partitions may cause the EM algorithm to converge to a local maximum of the likelihood function. We propose several simple and fast refinements based on data transformations and illustrate them through data examples.

• Publication year

  2015

• Venue

  Advances in Data Analysis and Classification

• Publication date

  2015-07-25

• Fields of study

  Mathematics, Computer Science, Medicine

• Identifiers
  DOI 10.1007/s11634-015-0220-zarXiv 1507.07114PMID 26949421PMCID PMC4776768
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar, PubMed

• No claims are published for this paper.

• No concepts are published for this paper.

• mclust 5: Clustering, Classification and Density Estimation Using Gaussian Finite Mixture Models

  2016cited by this paper
• Normal Mixture Modelling for Model-Based Clustering,Classification, and Density Estimation

  2015cited by this paper
• On comparing partitions

  2015influential reference
• Multivariate data analysis as a discriminating method of the origin of wines

  2015cited by this paper
• mclust Version 4 for R : Normal Mixture Modeling for Model-Based Clustering , Classification , and Density Estimation

  2012cited by this paper
• Initializing the EM algorithm in Gaussian mixture models with an unknown number of components

  2012cited by this paper
• Cluster Analysis: Everitt/Cluster Analysis

  2011cited by this paper
• Finite mixture models and model-based clustering

  2010cited by this paper
• Initializing Partition-Optimization Algorithms

  2009influential reference
• The EM Algorithm and Extensions: Second Edition

  2008cited by this paper
• Finite Mixture Models

  2008cited by this paper
• Model-based cluster and discriminant analysis with the MIXMOD software

  2006cited by this paper
• Variable Selection for Model-Based Clustering

  2006cited by this paper
• Choosing starting values for the EM algorithm for getting the highest likelihood in multivariate Gaussian mixture models

  2003influential reference
• Model-Based Clustering, Discriminant Analysis, and Density Estimation

  2002cited by this paper
• Finite Mixture Models

  2000cited by this paper
• Assessing a Mixture Model for Clustering with the Integrated Completed Likelihood

  2000influential reference
• Classification, 2nd Edition

  1999cited by this paper
• Algorithms for Model-Based Gaussian Hierarchical Clustering

  1998cited by this paper
• How Many Clusters? Which Clustering Method? Answers Via Model-Based Cluster Analysis

  1998cited by this paper
• A First Course in Multivariate Statistics

  1997cited by this paper
• The EM algorithm and extensions

  1996cited by this paper
• Gaussian parsimonious clustering models

  1995cited by this paper
• Model-based Gaussian and non-Gaussian clustering

  1993cited by this paper
• Finding Groups in Data: An Introduction to Cluster Analysis

  1990cited by this paper
• Algorithms for Clustering Data

  1988cited by this paper
• On the choice of starting values for the EM algorithm in fitting mixture models

  1988cited by this paper
• A Study of the Comparability of External Criteria for Hierarchical Cluster Analysis.

  1986cited by this paper
• ON THE CONVERGENCE PROPERTIES OF THE EM ALGORITHM

  1983cited by this paper
• Estimating the Dimension of a Model

  1978cited by this paper
• Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper

  1977cited by this paper
• Cluster Analysis

  1974cited by this paper
• Max-imum Likelihood from Incomplete Data

  1972cited by this paper

• Early Postpartum Metabolic Heterogeneity Among Women Who Progressed to Type 2 Diabetes After Gestational Diabetes: A Prospective Cohort

  2025cites this paper
• Bayesian Model Averaging with Diffused Priors for Model-Based Clustering Under a Cluster Forests Architecture

  2025cites this paper
• Mixtures of Quantile-Based Factor Analyzers

  2025cites this paper
• Mechanism-based subtyping in binge eating: understanding neurobehavioral heterogeneity across negative emotionality, approach behavior, and executive function

  2025cites this paper
• Data-driven investigation of traffic spatial correlation through functional data analysis

  2025cites this paper
• Model-based clustering approach for clustering of heart disease patients based on risk factors

  2025cites this paper
• Challenges and Adaptations of Model‐Based Clustering for Flow and Mass Cytometry

  2025cites this paper
• Autonomous Clustering by Fast Find of Mass and Distance Peaks

  2025cites this paper
• Machine learning reveals regime shifts in future ocean carbon dioxide fluxes inter-annual variability

  2024cites this paper
• A Model-Based Clustering Approach for Bounded Data Using Transformation-Based Gaussian Mixture Models

  2024influential citation
• Uncovering Patterns in Perceived Therapist Behaviour: A Latent Profile Analysis of Broaching Experiences Among Ethnic Minority Group Member Clients.

  2024cites this paper
• A mediated model of study choice profiles and person-environment fit with bachelor's program as predictors of academic success in higher education

  2024cites this paper
• Sparse Model-Based Clustering of Three-Way Data via Lasso-Type Penalties

  2023cites this paper
• A Gromov-Wasserstein Geometric View of Spectrum-Preserving Graph Coarsening

  2023cites this paper
• Improving Species Delimitation for Effective Conservation: A Case Study in the Endemic Maple-leaf Oak (Quercus acerifolia).

  2023cites this paper
• A Regime View of ENSO Flavors Through Clustering in CMIP6 Models

  2023influential citation
• Occupational resource profiles for an addressee orientation in occupational health management: a segmentation analysis

  2023cites this paper
• Latent profiles of emotion dysregulation among individuals with binge-spectrum eating disorders: Associations with eating disorder pathology.

  2023cites this paper
• Microbial communities form rich extracellular metabolomes that foster metabolic interactions and promote drug tolerance

  2022cites this paper
• ALGILANAN BAŞARILI YÖNETİCİ ÖZELLİKLERİ BAĞLAMINDA FARKLI SİYASİ EĞİLİME SAHİP İNSANLARA YÖNELİK STEREOTİPLERDE OLUŞAN ÖRTÜK PROFİLLER

  2022cites this paper
• Circuit analysis of the Drosophila brain using connectivity-based neuronal classification reveals organization of key information processing pathways

  2022cites this paper
• Seasonal circulation regimes in the North Atlantic: Towards a new seasonality

  2022cites this paper
• Sorensen-dice similarity indexing based weighted iterative clustering for big data analytics

  2022cites this paper
• Unsupervised expectation-maximization algorithm initialization for mixture models: A complex network-driven approach for modeling financial time series

  2022cites this paper
• Circuit analysis of the Drosophila brain using connectivity-based neuronal classification reveals organization of key communication pathways

  2022cites this paper
• Latent profiles of dietary restraint among individuals with binge-spectrum eating disorders: Associations with eating disorder symptom severity.

  2022cites this paper
• An integrated atlas of human placental development delineates essential regulators of trophoblast stem cells

  2022cites this paper
• A model-based clustering of expectation–maximization and K-means algorithms in crime hotspot analysis

  2022cites this paper
• Sputum alarmin levels delineate distinct T2 cytokine pathways and patient subgroups in asthma

  2022cites this paper
• Unobserved classes and extra variables in high-dimensional discriminant analysis

  2021cites this paper
• Engineering students' noncognitive and affective factors: Group differences from cluster analysis

  2021cites this paper
• Serum Fatty Acid Latent Classes Are Associated With Suicide in a Large Military Personnel Sample.

  2021cites this paper
• Group-wise shrinkage for multiclass Gaussian Graphical Models

  2021cites this paper
• Group-Wise Shrinkage Estimation in Penalized Model-Based Clustering

  2021cites this paper
• Partitioning the Galactic halo with Gaussian Mixture Models

  2021cites this paper
• Vine copula mixture models and clustering for non-Gaussian data

  2021cites this paper
• An artificial bee colony algorithm for mixture model-based clustering

  2020cites this paper
• Assessing Differences between Three Virtual General Chemistry Experiments and Similar Hands-On Experiments

  2020cites this paper
• Improved Initialization of the EM Algorithm for Mixture Model Parameter Estimation

  2020cites this paper
• Gaussian Mixture Model Based Classification Revisited: Application to the Bearing Fault Classification

  2020cites this paper
• Selecting the Number of Clusters K with a Stability Trade-off: an Internal Validation Criterion

  2020cites this paper
• Neuronal classification from network connectivity via adjacency spectral embedding

  2020cites this paper
• A stochastic block model for interaction lengths

  2020cites this paper
• Optimizing the Estimation of a Histogram-Bin Width—Application to the Multivariate Mixture-Model Estimation

  2020influential citation
• The Nature of Espeletia Species

  2020cites this paper
• Better than the best? Answers via model ensemble in density-based clustering

  2020cites this paper
• Using a Finite Mixture Model-Based Clustering Approach to Identify Move and Wait Strategies.

  2020cites this paper
• Model-Based Clustering

  2020cites this paper
• Model-based Classification

  2019cites this paper
• Bibliography

  2019cites this paper
• Introduction

  2019cites this paper
• Semi-supervised Clustering and Classification

  2019cites this paper
• Model-based Clustering: Basic Ideas

  2019cites this paper
• Model-based Clustering with Covariates

  2019cites this paper
• Non-Gaussian Model-based Clustering

  2019cites this paper
• Preface

  2019cites this paper
• Discrete Data Clustering

  2019cites this paper
• International Journal of Innovative Technology and Exploring Engineering (IJITEE)

  2019cites this paper
• Optimum average silhouette width clustering methods

  2019cites this paper
• How bettering the best? Answers via blending models and cluster formulations in density-based clustering

  2019cites this paper
• Multi-Subject Stochastic Blockmodels for Adaptive Analysis of Individual Differences in Human Brain Network Cluster Structure

  2019cites this paper
• Dealing with Difficulties

  2019cites this paper
• Model-Based Clustering and Classification for Data Science: With Applications in R

  2019cites this paper
• High-dimensional Data

  2019cites this paper
• Other Topics

  2019cites this paper
• Model-Based Clustering and Classification for Data Science

  2019cites this paper
• Variable Selection

  2019cites this paper
• A dynamic stochastic blockmodel for interaction lengths

  2019cites this paper
• Network data

  2019cites this paper
• Retrospective analysis of insulin responses to standard dosed oral glucose tests (OGTs) via naso-gastric tubing towards definition of an objective cut-off value

  2018cites this paper
• A transformation‐based approach to Gaussian mixture density estimation for bounded data

  2018cites this paper
• Anatomy of STEM teaching in North American universities

  2018cites this paper
• Determination of the Representative Socioeconomic Level by BSA in the Mexican Republic

  2018cites this paper
• Initialization of Recursive Mixture-based Clustering with Uniform Components

  2017cites this paper
• Estimation and testing problems in auditory neuroscience via clustering

  2017cites this paper
• Model-based clustering with sparse covariance matrices

  2017cites this paper
• Modelling the role of variables in model-based cluster analysis

  2017influential citation
• Practical Initialization of Recursive Mixture-Based Clustering for Non-negative Data

  2017cites this paper
• Constrained clustering with a complex cluster structure

  2016cites this paper
• mclust 5: Clustering, Classification and Density Estimation Using Gaussian Finite Mixture Models

  2016cites this paper
• clustvarsel: A Package Implementing Variable Selection for Gaussian Model-Based Clustering in R.

  2014cites this paper
• Subject Index

  1984cites this paper
• Title mclust 5 : Clustering , Classification and Density Estimation Using Gaussian Finite Mixture Models

  year unknowncites this paper