Event Recognition Challenges and Techniques

Abstract

Event recognition refers to the detection of events that are considered relevant for processing, thereby providing the opportunity to implement reactive measures. There are five research challenges of event recognition, namely, multiscale temporal aggregation, uncertainty, distribution, pattern learning, and event forecasting. The July 2014 Special Issue of ACM Transactions on Internet Technology introduces six articles on the challenges of event recognition. Approximate Semantic Matching of Events for The Internet of Things, PADUA: Parallel Architecture to Detect Unexplained Activities, Adaptive Speculative Processing of Out-of-Order Event Streams, Decentralised Fault-Tolerant Event Correlation, MCEP: A Mobility-Aware Complex Event Processing System, and Efficient Stream Provenance via Operator Instrumentation are the articles discussed in this issue.