Supervised Anomaly Detection in Uncertain Pseudoperiodic Data Streams
It addresses anomaly detection for sensor data in Web applications, but appears incremental as it builds on existing classification methods with added pre-processing.
The paper tackles anomaly detection in uncertain pseudoperiodic data streams by proposing a framework that pre-processes uncertainty, extracts features, and uses classification, achieving high accuracy on real datasets.
Uncertain data streams have been widely generated in many Web applications. The uncertainty in data streams makes anomaly detection from sensor data streams far more challenging. In this paper, we present a novel framework that supports anomaly detection in uncertain data streams. The proposed framework adopts an efficient uncertainty pre-processing procedure to identify and eliminate uncertainties in data streams. Based on the corrected data streams, we develop effective period pattern recognition and feature extraction techniques to improve the computational efficiency. We use classification methods for anomaly detection in the corrected data stream. We also empirically show that the proposed approach shows a high accuracy of anomaly detection on a number of real datasets.