Combining neural clustering with intelligent labeling and unsupervised Bayesian reasoning in a multiview context for efficient diachronic analysis

Abstract

To cope with the current defects of existing incremental clustering methods, an alternative approach for accurately analyzing textual information evolving over time consists in performing diachronic analysis. This type of analysis is based on the application of a clustering method on data associated with two, or more, successive periods of time, and on the study of the evolution of the clusters contents and of their mappings between the different periods. This paper propose a new unsupervised approach for dealing with time evolving information with is based on the combination of neural clustering and unsupervised Bayesian reasoning. The experimental context is related to the study of the evolution of research fields in scientific literature. © 2013 Springer-Verlag.