Optimized Stage Processing for Anomaly Detection on Numerical Data Streams - A Solution for DEBS Grand Challenge 2017
Emanuel Onica (Alexandru Ioan Cuza University of Iasi)
The DEBS Grand Challenge is a competition aimed at both research and industrial event-based systems organized yearly as part of the DEBS conference. The general scope is to implement a solution to a problem, which is evaluated and rated based on the criteria of correctness, throughput, and latency. The 2017 edition focused on the problem of automatic detection of anomalies for manufacturing equipment. This presentation reports the technical details of a solution focused on particular optimizations of the processing stages. These included customized input parsing, fine tuning of a k-means clustering algorithm and probability analysis using a lazy flavor of a Markov chain. The series of tweaks that will be discussed were implemented in a single node based solution, finally obtaining good performance results.
Emanuel Onica is a lecturer at the Faculty of Computer Science, Alexandru Ioan Cuza University, Iasi, Romania. He received his PhD from University of Neuchâtel, Switzerland, where he worked as a Scientific Collaborator between 2010 and 2014. He is currently coordinating the Horizon 2020 EBSIS Twinning project started in 2016. He was previously involved in the SRT-15 FP7 project and had collaborative work in the LEADS FP7 project. His research activity lies mostly in the area of distributed systems, where he is a co-author of a best paper at the DEBS 2013 conference.
Back to EBSIS Events section.