AURA Technology

The Cybula AURA technology is a range of patented solutions of pattern recognition and modelling tools for data mining and analytics on complex unstructured data.   These methods are based on binary neural network and associative memory techniques.  The methods are data agnostic meaning that we can, and do, work across a highly diverse range of application domains. 

AURAalert

Cybula's AURAalert pattern-matching technology is a highly novel approach to generating models of systems or assets for real-time condition monitoring.  Unique and highly intuitive in their application, they take a data driven approach to generating a dynamic model of the normal operating modes of a system.  In applications where it is difficult to produce a traditional statistical model of a system, this approach has huge benefits for detecting deviations from normal operating trends. The reference model is easy to build based on the assimilation of large volumes of raw data from the monitored system or asset.

AURA Pattern Match Search

Pattern recognition is at the heart of the AURA approach and the pattern match search engine provides an efficient, scalable method for detecting complex time-series events in data.   Where Big Data challenges abound, finding events of significance can appear to be a 'needle in a haystack' problem; the AURA search engine provides an analyst with tools to search deep into the data from a simple intuitive interface, and to build automated pattern search tasks to deploy on the data, even in real-time streaming situations.

AURA Transient Cluster

Comparing the performance of an asset or system with similar assets over time can provide powerful diagnostic insights for real business intelligence.  Few systems offer this capability due to the technical complexity of the task.   AURA Transient again simplifies this task and provides an analyst with a highly intuitive way to contrast and compare using a data driven modelling approach which can be specified and constructed in minutes with the tool.