Fast, accurate and explainable time series classification through randomization

Abstract

Time series classification (TSC) aims to predict the class label of a given time series, which is critical to a rich set of application areas such as economics and medicine. State-of-the-art TSC methods have mostly focused on classification accuracy, without considering classification speed. However, efficiency is important for big data analysis. Datasets with a large training size or long series challenge the use of the current highly accurate methods, because they are usually computationally expensive. Similarly, classification explainability, which is an important property required by modern big data applications such as appliance modeling and legislation such as the European General Data Protection Regulation, has received little attention. To address these gaps, we propose a novel TSC method – the Randomized-Supervised Time Series Forest (r-STSF). r-STSF is extremely fast and achieves state-of-the-art classification accuracy. It is an efficient interval-based approach that classifies time series according to aggregate values of the discriminatory sub-series (intervals). To achieve state-of-the-art accuracy, r-STSF builds an ensemble of randomized trees using the discriminatory sub-series. It uses four time series representations, nine aggregation functions and a supervised binary-inspired search combined with a feature ranking metric to identify highly discriminatory sub-series. The discriminatory sub-series enable explainable classifications. Experiments on extensive datasets show that r-STSF achieves state-of-the-art accuracy while being orders of magnitude faster than most existing TSC methods and enabling for explanations on the classifier decision.

Monitoring Flow-Forming Processes Using Design of Experiments and a Machine Learning Approach Based on Randomized-Supervised Time Series Forest and Recursive Feature Elimination

“…The r-STSF is a machine learning algorithm for time series data described in [ 8 ] as an advanced successor of the Time Series Forest [ 39 ] and the Supervised Time Series Forest [ 40 ]. In the area of time series classification, the r-STSF achieves the state-of-the-art performance of complex and computationally expensive algorithms such as HIVE-COTE (Hierarchical Vote Collective of Transformation-Based Ensembles, see [ 41 ]) while being several orders of magnitude faster [ 8 ]. A feature extraction algorithm, which generates a pool of interval features from a set of time series data, is a key aspect of r-STSF.…”

“…By implementing the feature extraction functionality autonomously (instead of using the complete r-STSF algorithm), the following advantages are achieved: A feature selection can be incorporated between the feature extraction and the training of models on the generated features (see Figure 14 ). The Extra Tree algorithm is an integral part of the r-STSF as defined by [ 8 ]. By detaching the feature extraction functionality from the rest of r-STSF, it becomes possible to train arbitrary other machine learning algorithms on the generated features.…”

“…By detaching the feature extraction functionality from the rest of r-STSF, it becomes possible to train arbitrary other machine learning algorithms on the generated features. In addition to the choice of learning algorithm, the choice of libraries and implementations used for it is also open (Cabello et al [ 8 ] use the Python library sklearn). In the present work, the machine learning platform H2O is used to train the models.…”

“…The approach includes an experimental design capable of providing the necessary data, as well as a procedure for preprocessing the data and extracting features that capture the information needed by the machine learning models to detect defects in the blank and the machine. Development of a feature extraction algorithm for multivariate time series classification, derived from the state-of-the-art algorithm r-STSF (see [ 8 ]), incorporating extensions for multivariate time series and metric target variables. …”

“…Development of a feature extraction algorithm for multivariate time series classification, derived from the state-of-the-art algorithm r-STSF (see [ 8 ]), incorporating extensions for multivariate time series and metric target variables.…”