DMX
Once you have built your data mining models, you can query it using DMX (Data Mining Extensions for SQL). As the name suggest, it extends the SQL syntax to allow you to query the mining models. For those that are familiar with SQL, it is easy to understand DMX.
E.g.
SELECT TOP 25 t.CustomerID
FROM CustomerChurnModel
NATURAL PREDICTION JOIN
OPENQUERY('CustomerDataSource', 'SELECT * FROM Customers')
ORDER BY PredictProbability([Churned],True) DESC
Several interesting constructs that pop up includes the PREDICTION JOIN and data mining specific functions like PredictProbability(...).
The Prediction Join allows you to probe a data mining model by supplying it with new case(s). Will be elaborating on this further in the next post. (Re-think what a join does in an RDBMS!)
E.g.
SELECT TOP 25 t.CustomerID
FROM CustomerChurnModel
NATURAL PREDICTION JOIN
OPENQUERY('CustomerDataSource', 'SELECT * FROM Customers')
ORDER BY PredictProbability([Churned],True) DESC
Several interesting constructs that pop up includes the PREDICTION JOIN and data mining specific functions like PredictProbability(...).
The Prediction Join allows you to probe a data mining model by supplying it with new case(s). Will be elaborating on this further in the next post. (Re-think what a join does in an RDBMS!)
0 Comments:
Post a Comment
<< Home