Speaker
Mr
Viktor Kotliar
(IHEP)
Description
An anomaly or an outlier in related multidimensional data can seriously affect on their analysis. Discarding anomalies or their ignoring may lead to incorrect results. A study was carried out by using the flexible PyOD tool which are designed for anomalies searching in multidimensional data. This study using various techniques or python software library modules provided by PyOD. The search goal was to discover anomalies in pair of indicators Ibat-TupsC on Symmetra UPS installed at IHEP IT center.
This paper presents research results that will be used to further development of the system for searching anomalies in engineering equipment of the IHEP IT center.
Primary author
Mr
Viktor Kotliar
(IHEP)
Co-author
Mrs
Ekaterina Popova
(IHEP)
