IPredictorClient::PredictMostLikelySegment
Use this method to predict the most likely segment of a population to which a user, the current case, belongs.
Definition
HRESULT IPredictorClient::PredictMostLikelySegment(IDictionary*dCurrentCase,VARIANT*pvarProbability,BSTR*pbstrSegment);
Parameters
dCurrentCase
[in] A pointer to the IDictionary interface that contains the current case.
pvarProbability
[out, optional] A pointer to a VARIANT used to return the membership probability associated with the segment name.
pbstrSegment
[out, retval] A pointer to a BSTR used to return the name of the segment most likely to contain the given case.
Return Values
This method returns an HRESULT indicating whether or not it completed successfully. See the Error Values section for more details.
Error Values
This method returns S_OK (0x00000000) to indicate success and standard COM HRESULT error values to indicate failure. For more information about standard COM errors, seeĀ Standard COM Errors. Additional information may be available using the global Err object, which can be accessed using the API function GetErrorInfo. In particular, the GetDescription method of the IErrorInfo interface may return a text description of the error.
Remarks
The Microsoft Research Cluster Engine Data Modeler groups similar users into segments. Similar segments are further grouped into higher order collections of segments, creating a hierarchy from which to analyze trends and discover hidden correlations.
This method returns the name of the segment most likely to contain the given case, and optionally, the probability that the user belongs to this segment.
Load the segment model using the LoadModelFromDB or LoadModelFromFile methods prior to using this method.
For more information about memory ownership issues related to COM property and method parameters, seeĀ Managing COM Parameter Memory.
The pvarProbability and pbstrSegment parameters contain valid data only if the method completes successfully.