An overview of clustering methods with guidelines for application in mental health research

被引：44

作者：

Gao, Caroline X. ^{[1
,2
,3
,7
]}

Dwyer, Dominic ^{[1
,2
]}

Zhu, Ye ^{[4
]}

Smith, Catherine L. ^{[3
]}

Du, Lan ^{[5
]}

Filia, Kate M. ^{[1
,2
]}

Bayer, Johanna ^{[1
,2
]}

Menssink, Jana M. ^{[1
,2
]}

Wang, Teresa ^{[5
]}

Bergmeir, Christoph ^{[5
,6
]}

Wood, Stephen ^{[1
,2
]}

Cotton, Sue M. ^{[1
,2
]}

机构：

[1] Univ Melbourne, Ctr Youth Mental Hlth, Parkville, Vic, Australia

[2] Orygen, Parkville, Vic, Australia

[3] Monash Univ, Sch Publ Hlth & Prevent Med, Dept Epidemiol & Preventat Med, Melbourne, Vic, Australia

[4] Deakin Univ, Sch Informat Technol, Geelong, Vic, Australia

[5] Monash Univ, Fac Informat Technol, Clayton, Vic, Australia

[6] Univ Granada, Dept Comp Sci & Artificial Intelligence, Granada, Spain

[7] Univ Melbourne, Orygen, Ctr Youth Mental Hlth, Locked Bag 10 35 Poplar Rd, Parkville, Vic 3052, Australia

来源：

PSYCHIATRY RESEARCH | 2023年 / 327卷

关键词：

Clustering; Cluster analysis; Machine learning; Unsupervised learning; Mental health research; CLASS GROWTH ANALYSIS; LATENT CLASS ANALYSIS; MIXTURE-MODELS; R-PACKAGE; COPHENETIC CORRELATION; HIERARCHICAL TAXONOMY; MAXIMUM-LIKELIHOOD; ANOMALY DETECTION; CROSS-VALIDATION; CLASSIFICATION;

D O I：

10.1016/j.psychres.2023.115265

中图分类号：

R749 [精神病学];

学科分类号：

100205 ;

摘要：

Cluster analyzes have been widely used in mental health research to decompose inter-individual heterogeneity by identifying more homogeneous subgroups of individuals. However, despite advances in new algorithms and increasing popularity, there is little guidance on model choice, analytical framework and reporting requirements. In this paper, we aimed to address this gap by introducing the philosophy, design, advantages/disadvantages and implementation of major algorithms that are particularly relevant in mental health research. Extensions of basic models, such as kernel methods, deep learning, semi-supervised clustering, and clustering ensembles are sub-sequently introduced. How to choose algorithms to address common issues as well as methods for pre-clustering data processing, clustering evaluation and validation are then discussed. Importantly, we also provide general guidance on clustering workflow and reporting requirements. To facilitate the implementation of different al-gorithms, we provide information on R functions and libraries.

引用

页数：28