Towards a cloud-based framework for online and integrated event detection

Abstract

Time series events detection relates to the study of techniques for detecting points in a series with special meaning which differs from the expected behavior of the dataset. In scenarios such as digital twins and IoT devices, there is natural generation and traffic of data in the cloud. Event detection is critical for timely decision-making. Since many methods for detecting events target different types selecting a suitable method makes the task more difficult. In this context, this article proposes a cloud-based framework called Harbinger Nimbus. The implementation was evaluated on the Microsoft Azure platform.