Mohammed Waleed Kadous
School of Computer Science & Engineering
University of New South Wales
Sydney, NSW 2052, Australia
waleed@cse.unsw.edu.au
Supervised classification is one of the most active areas of machine learning research. Most work has focused on classification in static domains, where an instantaneous snapshot of attributes is meaningful. In many domains, attributes are not static; in fact, it is the way they vary temporally that can make classification possible. Examples of such domains include speech recognition, gesture recognition and electrocardiograph classification. While it is possible to use ad hoc, domain-specific techniques for ``flattening'' the time series to a learner-friendly representation, this fails to take into account both the special problems and special heuristics applicable to temporal data and often results in unreadable concept descriptions. Though traditional time series techniques can sometimes produce accurate classifiers, few can provide comprehensible descriptions. We propose a general architecture for classification and description of multivariate time series. It employs event primitives to analyse the training data and extract events. These events are clustered, creating prototypical events which are used as the basis for creating more accurate and comprehensible classifiers. A minimal implementation of this architecture, called TClass, is applied to two domains, one real and one artificial and compared against a naïve approach. TClass shows great promise, particularly in comprehensibility, but also in accuracy.
Keywords: machine learning, classification, temporal classification, gesture recognition, time series.