Loom Systems takes digitized information in structured, unstructured, non-standard or uncommonly structured text format and structures it automatically.
By mathematically modeling how humans analyze such structures, Loom Systems fuses analytical skills with computational speed to simulate and enhance the entire data analysis cycle.
Loom Systems' solution ingests all types of machine data out-of-the-box, whether it’s structured or unstructured. Its automated log message pre-processing algorithms (ALPA) automate the reconstruction process of textual events, leveraging structure recognition methodologies to transform unstructured data to structured data.
Given any textual source or stream of events, the algorithm will reverse-engineer one or more structures common to data events. As part of this process, the algorithm will tell apart the original event from any peripheral data. Any textual log message (unstructured data) is consequently transformed into structured data.
Our solution takes each metric and tracks it. It learns its unique baseline and behavioral pattern over time to be able to detect anomalies and predict future trends.
It uses complex modules to determine whether a signal has shifted, as well as the type of shift. The signal types are distinguished and anomaly detection algorithms are tailored to fit them. Signals are thus automatically tracked in ways that complement their expected behavior.
Our solution enriches alerts with recommended resolutions from its Intelligent Insights Network (IIN). This DB contains a wide set of built-in recommended resolutions provided as part of the Loom Systems offering.
Moreover, the DB can ingest any existing proprietary knowledge within the organization to streamline the detection to resolution process of issues in proprietary applications within the system.
Lastly, as the IIN is adaptive, it continuously learns from the users and expands the breadth of recommended resolutions available.
