PredictorServiceSiteAdmin Object
Use this object to manage analysis models and model configurations.
ProgID: | Commerce.PredictorSiteAdmin (Externally creatable) |
COM Class Name: | PredictorServiceSiteAdmin |
COM Interface Name: | IPredictorServiceSiteAdmin |
Interface ID Constant: | IID_IPredictorServiceSiteAdmin |
Header File: | PredictorService.h |
Type Library Name: | Microsoft Commerce 2000 Predictor Service Type Library |
EXE Name: | PredServ.exe |
Threading Model: | Both |
In C++, use the IPredictorServiceSiteAdmin interface to access the properties and methods of the PredictorServiceSiteAdmin object.
Because this object is primarily intended for use within ASP script, access from C++ will not be common.
The properties of the PredictorServiceSiteAdmin object are shown in the following table.
Property | Type | Description |
lCaseCount | long | Retrieves the number of cases used by a model configuration.
This property is read-only. |
dModelInfo | IDictionary | Retrieves information on the specified model.
This property is read-only. |
SegmentLabels | VARIANT | Stores the segment labels for a segment model.
This property is read/write. |
The methods of the PredictorServiceSiteAdmin object are shown in the following table.
Method | Description |
DeleteModel | Deletes the specified predictor model. |
DeleteModelConfig | Deletes the specified model configuration. |
GenerateSegmentList | Generates a list of members of a segment. |
Init | Initializes the server. |
RenameModel | Renames a predictor model. |
RenameModelConfig | Renames a model configuration. |
Remarks
An analysis model summarizes dependency relationships among attributes found in data. A model configuration contains information used to build the analysis model.
Two types of analysis models are available. A prediction model is used to produce a ranked list of recommendations or to fill in a missing user profile attribute. A segment model is used to group similar users into segments of a population.
The PredictorServiceAdmin object contains lists of the available models and model configurations. The PredictorClient object loads the appropriate model and makes the actual predictions. The generation of new models is controlled by the PredModelBuilder object.