When you create a new mining model with the Data Mining Wizard, It seems that the wizard does not detect correctly the content type of attributes if your local settings are other than US English. During the wizard, there is
a button called “Detect” in one of the pages; if you click on it, it detects the content type better (still not perfect). For example, I tested Naïve Bayes algorithm without clicking on the Detect button, and content of all attributes (the Content property) was Discretized, with empty DiscretizationBucketCount and DiscretizationMethod properties; these settings make no sense. Therefore, it is logical the algorithm does not find any interesting dependency. For Decision Trees algorithm, I got many columns with continuous content, although the content was actually discrete. In this case I got a regression tree instead of decision tree as I expected.You can change the Content property of attributes after you create a structure and models with the wizard. You can do it in the Designer window, using the Mining Structure tab (the first tab). Click on each column and check the Content, DiscretizationBucketCount and DiscretizationMethod properties. In addition, if your columns were detected as continuous and you use Decision Trees algorithm, you have to change the ModelingFlags of the attribute in the model as well. In the Designer window, check the properties of the attributes used in a model. Be sure to click on the column in the model and not in the structure area, and then clear the check box near the REGRESSOR flag of the ModelingFlags property, as shown in the picture bellow.

Dejan Sarka