Using Jaccard Distance Measure for Unsupervised Activity Recognition with Smartphone Accelerometers

The rapid popularity of smartphones has led to a growing research interest in human activity recognition (HAR) with the mobile devices. Accelerometer is the most commonly used sensor of smartphone for HAR. Most supervised HAR methods have been developed. However, it is very difficult to collect the annotated or labeled training data for HAR. So, developing of effective unsupervised methods for HAR is very necessary. The accuracy of the unsupervised activity recognition can be greatly affected by feature extraction methods and distance measures. Although Euclidean distance measure is commonly used in activity recognition, it is not suitable for measuring distance when the number of features is very large, which is usually the case in HAR. Jaccard distance is a distance measure based on mutual information theory and can better represent the differences between nonnegative feature vectors than Euclidean distance. In this work, the Jaccard distance measure is applied to HAR for the first time. In the experiments, the results of the Jaccard distance measure and the Eucildean distance measure are compared, using three different feature extraction methods. Two different evaluation methods are used to comprehensively analyze the final results: (a) C-Index before clustering, (b) FM-index after using five different clustering methods which are Spectral Cluster, Single-Linkage, Ward-Linkage, Average-Linkage, and K-Medoids. Experiments show that, almost for every combination of the feature extraction methods and the evaluation methods, the Jaccard distance is consistently better than the Euclidean distance for unsupervised HAR.