After you train a data mining model in Microsoft SQL Server Analysis Services, you can explore the model to look for interesting trends. Because the results of mining models are complex and can be difficult to understand in a raw format, visually investigating the data is often the easiest way to understand the rules and relationships that algorithms discover within the data.
Exploring a data mining model lets you understand the behavior of the model before you deploy it. Each algorithm that you use to build a model returns a different type of results. Therefore, Analysis Services provides a separate viewer for each algorithm. When you browse a mining model in Business Intelligence Development Studio, the model is displayed on the Mining Model Viewer tab of Data Mining Designer, using the appropriate viewer for the model.
The viewer for each algorithm type also contains the Microsoft Generic Content Viewer. This generic viewer displays content that is common across all algorithms. The generic viewer contains information about the patterns and statistics that Analysis Services captures during the analysis, the characteristics and probabilities for individual nodes, regression formulas, and other technical details.
In This Section
The following topics describe the viewers that are provided in Analysis Services.
Describes the Microsoft Neural Network Viewer. This viewer displays mining models that are built with the Microsoft Neural Network algorithm, including models that use the Microsoft Logistic Regression algorithm.