I'm going now to continue from my first article "A neural Network in SQL Server - case study" where I showed the resulting data after I programmed a neural network (NN) in SQL Server. If you go back there, you might remember that I discovered that I needed first to better understand what variables of the data set were important for the prediction of rain for the next day, so I created a decision tree in Analysis Services and it gave me the idea of what variables to use for the input in the SQL neural network. Later, I also compared the results with the NN created in Analysis Services using the same data source.
In this article I will show you the process to create both, a decision tree and a neural network in Analysis Services, how to understand the information you'll get from them, and how to use it to make decisions or understand the meaning of your data.
The article continues at