Now Get Explanations for Anomaly Scores with Amazon Kinesis Analytics Anomaly Detection
Amazon Kinesis Analytics allows you to detect anomalies on streaming data in real time. Today, we launched two new features that provide explanations of the anomalies, making it easy for you to perform root cause analysis. You can learn which fields in your data lead to high anomaly scores and identify trends associated with the anomaly.
Kinesis Analytics uses Random Cut Forest algorithm to analyze one or more numeric fields and generate scores to identify anomalies in data streams. When records in your data stream have a large number of fields, it can be difficult to manually determine which fields lead to high anomaly scores, especially when the data is large, moves fast, and changes frequently. Kinesis Analytics now provides real-time explanations of the anomaly scores using attribution and directionality. Attribution explains the contribution of input fields to the overall score, and directionality provides information about the trends such as dips and spikes in each field. For more information and sample code, see Random Cut Forest with Explanations in the Amazon Kinesis Analytics SQL Reference.
Kinesis Analytics is the easiest way to process streaming data in real time with SQL without having to learn new programming languages or processing frameworks. Kinesis Analytics enables you to query streaming data or build entire streaming applications using SQL, so that you can gain actionable insights and respond to your business and customer needs promptly. Kinesis Analytics is available in the US East (N. Virginia), US West (Oregon), and EU (Ireland) regions.