Distance Ratio-based Weighted Rank Outlier Detection on Wearable Health Data

In the context of the popularity of smart wearable devices, aiming at the problem that the health data collected by the sports bracelet has unknown abnormal data which is related to specific disease, a method for detecting abnormal data of wearable health data based on Distance Ratio and Weighted Rank is proposed. Firstly, the t-Distributed Stochastic Neighbor Embedding is used to extract the features of the data set and enhance the local structure of the data. Then the distance ratio and weighted rank is used to replace the local reachable density in the Local Outlier Factor algorithm. A new algorithm for calculating the outlier factor is proposed which is called Distance Ratio-based Weighted Rank Outlier Factor (DRWROF) algorithm. Finally, the accuracy of the algorithm is verified by the simulation experiment on the UCI standard data set, meanwhile, an experimental analysis is performed on the actual activity data which is collected by sports bracelets. The experiment results show that it is suitable for the detection of outliers in the complex and diverse behaviors of different bracelet wearers in the actual dataset.