Problem: How to make use of a
data mining model.
Solution: Data
Mining Model Viewers
After the data mining structure, has been created, 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.
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 SQL Server Data Tools
(SSDT), the model is displayed on the Mining Model Viewer tab of Data Mining Designer,
using the appropriate viewer for the model.
How to Use the Model Viewers
First you select the mining model and then you select a
viewer. Each model always has two viewers available: a custom viewer, which can
include multiple tabs, and the generic viewer.
Depending on the type of the model that you selected, you
will see very different options for exploring the model. The custom viewers
associated with each model type are tailored to the algorithm that you used to
create the selected data mining model. Each custom viewer has a variety of
tools and dialog boxes for helping you explore the statistics and patterns in
the model, view charts, or interactively work with probability thresholds or
filter out items by name.
Example:
In this example, let’s see how we can use the data mining
model that has been created, please go through this article to see how we can
create a data mining structure. datamining
structure
Step 1: open the solution from visual studio and connect to
the mining structure and open the tab of data mining model viewer.
Step 2: toggle the mining model and viewer to change the
algorithm for using the data mining to get appropriate view in the results
window.
Results:
This is the Microsoft generic content viewer.
This is the Microsoft tree viewer where the probability is calculated based on the tree structure and decision trees.
0 comments:
Post a Comment