Microsoft Technology Licensing, LLC

Panagiotis Periorellis

Eldar Akchurin

Lev Novik

Olivier Nano

Tihomir Tsvetkov Tarnavski

Ivo Jose Garcia Dos Santos

Alexander Armin Spengler

Thomas Minka

Alexandre De Baynast

Vitaly V Filimonov

Dmitry Starostin

Aleksandr Klimov

The formulation of transformations on one or more input event streams to generation one or more output event streams. Accordingly, the transformations may be considered to be a query on the original input event stream(s). The event query includes event stream source representations representing an input event stream available in a particular execution context. The event query also includes a transformation module identifying the transformation set to be performed on the input event streams in the execution context. Once the query is properly formed, an execution module may then cause the transformations to be executed upon the designated input event stream(s) to generate output event streams.

Event stream transformations

Examining time series sequences representing performance counters from executing programs can provide significant clues about potential malfunctions, busy periods in terms of traffic on networks, intensive processing cycles and so on. An unsupervised anomaly detector can detect anomalies for any time series. A combination of known techniques from statistics, signal processing and machine learning can be used to identify outliers on unsupervised data, and to capture anomalies like edge detection, spike detection, and pattern error anomalies. Boolean and probabilistic results concerning whether an anomaly was detected can be provided.

Unsupervised anomaly detection for arbitrary time series

Growing community of inventors

Munich, Germany

Panagiotis Periorellis