Introduction
Cogynt Authoring is a web-based graphical tool that makes it possible to create visual representations of event patterns without needing to code or program. Cogynt Authoring facilitates various tasks to generate a model that then can be deployed to a Cogynt-powered Apache Flink cluster.
About This Guide
This guide is intended for data scientists or other subject matter experts who need to turn data sources into meaningful alerts and notifications for analysts.
After reading the guide, you should be able to create event patterns to match or filter against your data sources. In turn, Cogynt will inform your analysts when one of your event patterns is matched.
Cogynt Operating Principles
Cogynt operates by the principle of hierarchical complex event processing (HCEP). HCEP is a process by which large amounts of noisy or disorganized data are organized into coherent patterns. These patterns then make it possible to look for evidence of particular events.
Cogynt can create and process many such patterns–including highly complex ones–across a wide array of data sources, granting analysts large-scale monitoring and analysis capabilities. Cogynt Authoring's graphical interface facilitates the creation of event patterns, removing the need to code or program them.
Example
Suppose that your data sources observe the following events:
- Ringing bells
- A man in a tuxedo
- A woman in a wedding dress
- A collection of wrapped presents
Furthermore, suppose that these events are observed within the same geographical area within less than one hour of each other.
None of these observations are meaningful on their own. However, when combined, they indicate that a wedding is likely taking place.
Cogility offers training in HCEP concepts and strategy to help customers develop sophisticated, high-performance event patterns. To learn more about the available trainings, contact your Cogility sales representative.