Latent Tree Analysis
This provides a methodological enhancement for researchers in social sciences, medicine, and machine learning, though it appears incremental as an overview and extension of existing concepts.
The paper tackles the problem of modeling correlations among random variables using a tree of latent variables, proposing it as an improvement over latent class analysis to identify homogeneous subgroups in fields like social sciences and medicine, with applications in cluster analysis, topic detection, and deep probabilistic modeling.
Latent tree analysis seeks to model the correlations among a set of random variables using a tree of latent variables. It was proposed as an improvement to latent class analysis --- a method widely used in social sciences and medicine to identify homogeneous subgroups in a population. It provides new and fruitful perspectives on a number of machine learning areas, including cluster analysis, topic detection, and deep probabilistic modeling. This paper gives an overview of the research on latent tree analysis and various ways it is used in practice.