Using common statistical methods to model what is normal is a fairly straightforward task for systems with simple data patterns, but building more complex systems can present a considerable challenge.
The task was to find unusual, rarely occurring events or data for which little is known in advance. Anomalies are defined not by their own characteristics but in contrast to what is normal and they spoiled the regular processing cycle.
We've been able to build a highly accurate detector of anomalies. It helped to: (a) speed up a regular processing algorithms, (b) find algorithm defects, (c) increase system stability, and (d) create a customized processing routine which takes care of anomalies.
We've provided an effective way of to measure similarity between samples using generative models for likelihood estimation.
Probabilistic graphical models
unsupervised sound segmentation
Hi, we are Sciforce - a company where the integration of various branches of science builds up a powerful force to create robust software solutions. Working at the intersection of Computer Science with other technical, natural and humanitarian sciences let us go beyond traditional IT services and become both technical and scientific forces to our customers.