Multivariate Time Series Classification using Dilated Convolutional Neural Network
This addresses the problem of feature extraction in time series classification for domains like activity recognition, but it is incremental as it adapts existing CNN techniques.
The paper tackled multivariate time series classification by using a dilated convolutional neural network to automatically extract features, achieving results as effective as hand-crafted features on human activity recognition datasets.
Multivariate time series classification is a high value and well-known problem in machine learning community. Feature extraction is a main step in classification tasks. Traditional approaches employ hand-crafted features for classification while convolutional neural networks (CNN) are able to extract features automatically. In this paper, we use dilated convolutional neural network for multivariate time series classification. To deploy dilated CNN, a multivariate time series is transformed into an image-like style and stacks of dilated and strided convolutions are applied to extract in and between features of variates in time series simultaneously. We evaluate our model on two human activity recognition time series, finding that the automatic features extracted for the time series can be as effective as hand-crafted features.