Covid-19 News Clustering using MCMC-Based Learing of finite EMSD Mixture Models

With the growth of social media information on the Web, performing clustering on different types of data is a challenging task. Statistical approaches are widely used to tackle this task. Among the successful statistical approaches, finite mixture models have received a lot attention thanks to their...

Full description

Saved in:
Bibliographic Details
Main Authors: xuanbo su, Nizar Bouguila, Nuha Zamzami
Format: Article
Language:English
Published: LibraryPress@UF 2021-04-01
Series:Proceedings of the International Florida Artificial Intelligence Research Society Conference
Online Access:https://journals.flvc.org/FLAIRS/article/view/128506
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:With the growth of social media information on the Web, performing clustering on different types of data is a challenging task. Statistical approaches are widely used to tackle this task. Among the successful statistical approaches, finite mixture models have received a lot attention thanks to their flexibility. There are already many finite mixture models to cope with this task, but the Exponential Multinomial Scaled Dirichlet Distributions (EMSD) has recently shown to attain higher accuracy compared to other state-of-the-art generative models for count data clustering. Thus, in this paper, we present a Bayesian learning method based on Markov Chain Monte Carlo and Metropolis-Hastings algorithm for learning this model parameters. This proposed method is validated via extensive simulations and comparison with multinomial based mixture models.
ISSN:2334-0754
2334-0762