I am using SQL Server 2019 (Enterprise Edition) and looking for enhancing our SSAS model (MD) to provide foresting capabilities. Which are the options I have?
I saw that data mining feature was deprecated since SQL Server 2017. The next option is machine learning?
Any clue or suggestion to show this predictive data in Power BI is welcomed.
There isn't a replacement for the data mining tools in SSAS, it best to shift that back to the SQL Server DB layer, and then integrate the results into the cube.
If you are running SQL Server Enterprise 2019, your best option for machine learning and forecasting is the SQL Server Machine Learning Service, which can run R or Python work loads. It is limited in the packages you can use, but can meet most basic Machine Learning workloads.
The other option is to use something like the Azure Machine Learning suite of tools which you can connect to your on-prem SQL Server via a gateway. Another cloud option would be use Azure Databricks which can run a wide range of ML tasks Tensorflow, PyTorch, Mlib, MLFlow and other libraries. Depending on the amount of data, you may be able to use the ML tools in Power BI Dataflows, but that will require Embedded or Premium to integrate AutoML functions.