SAVE - Smart Alarms & Verified Events

Auf einen Blick

Projektleiter/in : Prof. Dr. Kurt Stockinger
Projektteam : Katrin Affolter, Prof. Dr. Martin Braschler, Jürg Deneke, Pascal Hulalka, Peter Monte, Markus Proelss, Ana-Claudia Sima, Jan Stampfli, Emil von Wattenwyl
Projektstatus : abgeschlossen
Drittmittelgeber : KTI
Projektpartner : Sitasys
Kontaktperson : Kurt Stockinger

Beschreibung

False alarms are a major cost driver for owners of alarm systems. To remedy this problem, we develop a novel alarm verification service by leveraging the power of an alarm data warehouse. By analyzing live streams of alarms, comparing them with historic alarms and enriching them with information from open data, our system learns how to identify false alarms, which helps human responders in their decision about triggering costly intervention forces or not.

In this project we will apply modern big data technology such as Spark SQL und Spark Streaming as well as state-of-the-art information retrieval and machine learning algorithms to solve this challenge with our industry partner Sitasys – one of the international leaders in secure alarm transmission.

Publikationen

Sima, Ana-Claudia; Stockinger, Kurt; Affolter, Katrin; Braschler, Martin; Monte, Peter; Kaiser, Lukas,

2018.

A hybrid approach for alarm verification using stream processing, machine learning and text analytics [Paper].

In:

Proceedings of the 21st International Conference on Extending Database Technology.

EDBT 2018, Vienna, Austria, 26-29 March 2018.

Association for Computing Machinery.

Verfügbar unter: https://doi.org/10.21256/zhaw-3487
Stampfli, Jan; Stockinger, Kurt,

2016.

Applied data science : using machine learning for alarm verification : a novel alarm verification service applying various machine learning algorithms can identify false alarms.

ERCIM News.

107, S. 10.

Verfügbar unter: https://doi.org/10.21256/zhaw-3785

Zurück