Anomaly Detection

Created with Sketch.

Get the right alarm at the right moment

Because data is so large and constantly changing, it is difficult to identify critical business issues by allowing people to look into the dashboard and apply simple rules to enormous data. Metatron Anomaly automatically identifies the trends and periodicity of your data, recommends the optimal anomaly detection algorithm, and lets you know when you need it. From business users who don’t know much about machine learning to knowledgeable data scientists, everyone is free to machine-learning with Metatron Anomaly.


Predictive model recommendations
Automatically recommends a prediction algorithm to match the characteristics of your data. Be aware of the problem before any unusual network activity occurs.

Alarm & Report
Identify critical business issues at the right moment. Immediately check for unusual situations and report them immediately to stakeholders.

Dashboard Analysis
Share your important data with your team, freely configured with charts and dashboards. Metatron Anomaly works seamlessly with Metatron Discovery, a self-service big data discovery tool.

Link to other systems
Supports seamless linkage with external analysis systems for applying the latest data analysis technology.

Automatic modeling of complexity in the real world

Metatron Anomaly learns the type of behavior of the data on its own. There is no need to build a statistical model or estimate thresholds as empirical values to understand future problem situations. Metatron Anomaly allows you to more easily identify problem situations. When you register the data you want to analyze and select the metrics you want to monitor, Metatron Anomaly recommends a model that fits your data. Compare the models and select the most convincing model. Now the machine learning model recommends thresholds and begins to detect anomalies that escape.

Detecting anomalous patterns in advance

Metatron Anomaly’s machine learning model extends the horizon of discoverable data coverage. As soon as a particular metric begins to deviate from the standard value, it can be detected immediately, or if it falls outside the “normal” profile range for the system, and can learn the normal value of a given group and capture unusual events that deviate from it.

Find insights on problem situations and improve efficiency

Check the most important information from the registered equipment, system and sensor data. Registered models and rules let you see at a glance the history of alarms that have occurred so far. Establish countermeasures for high-frequency failing systems and ensure that appropriate action has been taken. Prepare new measures through accumulated records.

Analyze anomalies with Metatron Discovery

Want to dig deeper into the data where irregular alarms occur? Do you want to display different charts such as radar charts, map charts, heatmaps, etc. in conjunction with other related data? Metatron Anomaly uses Metatron Disocvery’s advanced visualization capabilities to analyze data directly without unnecessary additional loading of data.