Multivariate Time Series Classification with Temporal Abstractions

Iyad Batal, Lucia Sacchi, Riccardo Bellazzi, Miloš Hauskrecht

2009 · 55 citations · 14 references

Concepts

TL;DR

The growing volume of complex temporal datasets has driven the extension of classical machine learning to time‑series, with multivariate time‑series classification being particularly challenging due to the vast number of potentially useful temporal features. The study develops a temporal abstraction framework to generate multivariate time‑series features for classification. The STF‑Mine algorithm automatically mines discriminative temporal abstraction patterns and uses them to train a classification model. Experiments on synthetic and real medical data show that the approach yields more accurate classifiers for time‑series datasets.

Abstract

The increase in the number of complex temporal datasets collected today has prompted the development of methods that extend classical machine learning and data mining methods to time-series data. This work focuses on methods for multivariate time-series classification. Time series classification is a challenging problem mostly because the number of temporal features that describe the data and are potentially useful for classification is enormous. We study and develop a temporal abstraction framework for generating multivariate time series features suitable for classification tasks. We propose the STF-Mine algorithm that automatically mines discriminative temporal abstraction patterns from the time series data and uses them to learn a classification model. Our experimental evaluations, carried out on both synthetic and real world medical data, demonstrate the benefit of our approach in learning accurate classifiers for time-series datasets.

References

14