A Bayesian Hybrid Approach to Unsupervised Time Series Discretization

Discretization is a key preprocessing step in knowledge discovery to make raw time series data applicable to symbolic data mining algorithms. To improve the comprehensibility of the mined results, or to help the induction step of the mining algorithms, in discretization, it is natural to prefer having discrete levels which can be mapped into intuitive symbols. In this paper, we aim to make smoothing of the data points along with the time axis, and make binning or clustering at the measurement axis. In particular, we propose a hybrid discretization method based on variational Bayes, in which the output of one discretization method is smoothly exploited as hyper parameters of another probabilistic discretization model such as a continuous hidden Markov model. The experiments with artificial and real datasets exhibit the usefulness of this hybrid approach.