New for machine learning in 7.12: Retention policy for transforms and Hyperparameter importance.
Extracting new insights from your Elasticsearch data is as simple as clicking a button - making machine learning truly operational. Sure, we love building a good algorithm, but you don't have to. Machine learning muscle is baked right into Elasticsearch and Kibana for an experience that's both powerful and performant.
If your data is in Elasticsearch, it's ready for machine learning. The Elastic Stack processes data upon ingest, ensuring that you have the metadata you need to identify root causes or add context to any event.
Not sure which jobs make sense for a new data set? We've done the work for you, finding the algorithms that will work at scale. Built-in tools like Data Visualizer help you find the
droids jobs you're looking for and identify fields in your data that would pair well with machine learning.
Unsupervised machine learning with Elastic helps you find patterns in your data. Use time series modeling to detect anomalies in your current data and forecast trends based on historical data. Wondering how your metrics are stacking up? Use outlier detection to zoom in on data points that stray from the rest.
Apply classification, regression, and outlier detection to your data for an end-to-end workflow experience across a wide range of use cases. Use continuous index transforms to convert an application logs index into a user-centric activity view, and build a fraud detection model using classification. Then use the inference ingest processor to apply your models to your incoming data at ingest time without ever leaving Elasticsearch.
Whether you're new to machine learning or are a seasoned data scientist, creating a machine learning job just makes sense — like catching unusually slow response times for your app directly in the APM app or discovering unusual behavior in the SIEM app. It might not be as simple as ordering a pizza online, but it’s getting pretty close.