Posted On: Mar 29, 2021

Today, we are announcing the general availability of ML Detect, a new feature of Amazon IoT Device Defender that automatically detects device-level operational and security anomalies across your IoT fleet by learning from your past data. Now, in addition to being able to manually set static alarms with Amazon IoT Device Defender Rules Detect, you can use machine learning to automatically learn your fleet's expected behaviors so that you don’t need an in-depth understanding of how your devices act across a range of metrics to get started.  

With ML Detect, you can monitor device behaviors based on 13 cloud-side and device-side metrics such as authorization failures, messages sent, and listening TCP port count. Additionally, ML Detect automatically updates the expected behaviors based on new data trends of your fleet caused by seasonality or other changing factors. When an anomaly is detected, you can respond by choosing a built-in mitigation action such as quarantining a device.

ML Detect is available in all Amazon Web Services regions where Amazon IoT Device Defender is available, including the Amazon Web Services China (Beijing) Region, operated by Sinnet and the Amazon Web Services China (Ningxia) Region, operated by NWCD. To learn more about the feature, please see the launch blog post and Amazon IoT Device Defender developer guide.