Summary
Historically, application log and telemetry data were used merely for support, diagnostics, and live site monitoring. With the reduced costs in storage and compute resources, this data can now be retained for longer for richer analysis. ML brings a huge opportunity to use log and telemetry data to gain deep, meaningful insights with real business impact.
In this chapter, you learned how to introduce ML models to your Data Explorer pool data using AutoML. We explored how to create Azure Machine Learning workspaces, how to configure the linked service to connect your Azure Synapse workspace to Azure Machine Learning, and finally, how to retrieve the best model from an AutoML experiment.
Next, you learned about other means to bring ML into your Azure Synapse projects. We looked at using pre-trained models to make predictions using Azure Cognitive Services, using KQL plugins to find patterns in data, and training ML models using Apache Spark MLlib.
Lastly, we briefly discussed...