Contextual Outlier Detection in Sensor Data Using Minimum Spanning Tree Based Clustering

Outlier detection is a fundamental data science task with a wide range of applications, including fraud detection, network security, environmental monitoring, agricultural management, and public health surveillance. The detection of outliers in sequential data has been widely studied in the data mining field, and many techniques have been developed for performing this task. However, most of the techniques are supervised in nature and thus need labeled data to train the model. In this paper, we introduce a graph-based approach to detecting contextual outliers in time-series sensor data. The algorithm presented offers more flexibility and requires less information about the nature of the analyzed data than previous approaches. Detection accuracy with this approach on different sets of wireless sensor data exceeded 90%, indicating that the proposed algorithm is effective for detecting outliers in diverse types of data in wireless sensor networks.