A novel similarity measure for time series data with applications to gait and activity recognition

In this abstract, we propose a novel approach to modeling time-series for the purpose of comparing segments of data in order to classify activities based on accelerometer sensor data. Our approach consists of producing an ensemble of simple classifiers that can be built and can classify new data efficiently. We present empirical results from an implementation of our algorithm running on a mobile phone, demonstrating the efficiency and performance of our technique on real-world data. Our algorithm is able to identify individuals based on their gait, and can be used in a semi-supervised setting to label large data sets using a small number of labeled examples. Our method can also be used in an unsupervised setting to visualize time-series data, for example, to identify the number of different activities that occur in an unlabeled data set.