Mª Concepción Bielza Lozoya: 'Modeling multivariate time series with Bayesian networks'

Mª Concepción Bielza Lozoya, Universidad Politécnica de Madrid

Abstract:

This talk will describe how Bayesian network models can cope with multivariate temporal data. After a brief introduction to Bayesian networks in static domains, discrete-time versions will be presented, with special focus on dynamic Bayesian networks. The more recent continuous time Bayesian networks and their supervised classification counterparts, both in uni- and multi-dimensional settings, will be then explained. Real-world examples from the industry will illustrate all models.