IoT-Gadgets-Logo-272-90

ARTIK introduces Rules on its machine learning cloud services

By
Sumit
 - 
Dec 13, 2017

Samsung ARTIK has introduced 'Rules' on ARTIK cloud services that utilize machine learning, the company announced on Monday. With the introduction of the new technology, ARTIK cloud services can now trigger Rules from actual device data. Furthermore, the cloud services can now predict data values as well as detect any anomaly in data values.

Users can create the data prediction and anomaly detection conditions in the web interface or using the Rules Application Programming Interface (APIs). Samsung says it is only the beginning of machine learning on ARTIK cloud services. "We wanted to show you what it can already bring to your Rules," a press release by the company stated.

 

What is machine learning?

Machine learning is the technology that gives computers the ability to make data predictions, driven by sample data usage/input patterns. On Samsung ARTIK cloud services, machine learning trains one of two model types that learn your device’s data usage.

  • Prediction Model - Predicts future data values based on current device data.
  • Anomaly Detection Model - Identifies anomalies in device data.

 

Apply machine learning to a Rule

You can include one of the above two models as a condition in a Rule. When you create a Rule from the My ARTIK Cloud interface, you have to select a device and then a field to use in the IF condition. This defaults to the Actual Value of the field.

 

Steps to create a prediction condition

  • Click the drop-down arrow next to Actual Value
  • Choose "use the predicted value"
  • Set the delta time
  • Click "Save"

 

Steps to create an anomaly detection condition

  • Click the drop-down arrow next to Actual Value
  • Choose "use the actual value"
  • Select “an expected value” or “an unexpected value” in the operator field
  • Set the confidence level. High confidence level identifies many anomalies whereas Low identifies less anomalies.

 

Please note that these models take some time to be trained. The conditions will be triggered only when machine learning is complete.

IoT-Gadgets-Logo-272-90

About us

IoT Gadgets is dedicated to bring you all the Internet of Things IoT news that pertains to gadgets. Simple. We love for you to join us on this journey.

Contact us: [email protected]

FOLLOW US

linkedin facebook pinterest youtube rss twitter instagram facebook-blank rss-blank linkedin-blank pinterest youtube twitter instagram