You are here: Symbol Reference > Classifier Namespace > Classes > TKNearestNeighbors Class > TKNearestNeighbors Members
Data Miner
ContentsIndex
PreviousUpNext
TKNearestNeighbors Members

The following tables list the members exposed by TKNearestNeighbors.

 
Name 
Description 
 
Called by LearnData method and K-NN to browse through the learn dataset.  
 
Called by LearnData method and K-NN to browse through the learn data set.  
 
Called by LearnData method and K-NN to restore the current position of the dataset.  
 
Called by LearnData method and K-NN to save the current position of the dataset.  
 
Name 
Description 
 
Copies the attribute Enabled fields from Source.  
 
Returns an array of attribute indexes sorted by quality.  
 
Determine the most probable class to which the example with values of discrete attributes in DiscreteRecord and values of real valued attributes in FloatRecord belongs to.  
 
Determine the response of all classe for the example with values of discrete attributes in DiscreteRecord and values of real valued attributes in FloatRecord.  
 
Performs the classification on the test data, by calling the OnClassifyTest event.  
 
Get the index of the class with Name.  
 
Get the name of the class at Index.  
 
Clear all learned data and all class descriptions.  
 
Create the component.  
 
This is Destroy, a member of class TKNearestNeighbors. 
 
Disable real valued and discrete attributes for all classes.  
 
Disable discrete attributes for all classes.  
 
Disable real valued attributes for all classes.  
 
Enable real valued and discrete attributes for all classes.  
 
Enable discrete attributes for all classes.  
 
Enable real valued attributes for all classes.  
 
Insert a record holding in to the K nearest neighbors array.  
 
Learn a new record belonging to class with ClassName.  
 
Call this method to perform the learn operation on the learn data.  
 
Learn a new record belonging to class with ClassIndex.  
 
Load the component from file named FileName.  
 
Load the component from stream.  
 
Returns the index of the class, which has the highest count of examples and thus the highest prior probability, with ClassIndex.  
 
Pruning disables some attributes, to improve classification accuracy.  
 
Pruning disables some attributes, to improve classificaiton accuracy.  
 
Performs pre-pruning and post-pruning and enables only those attributes giving best classification accuracy towards the test dataset.  
 
Reset all learned data including attribute weights and attribute Enabled fields, but keep the class descriptions.  
 
Save the component to file named FileName.  
 
Save the component to stream.  
 
Reset all learned data except attribute weights and attribute Enabled fields and keep the class descriptions.  
 
Sort an array of TIndexRecords.  
 
Name 
Description 
 
Set the value at index to True/False, to enable/disable the corresponding attribute.  
 
A list of recognized classes.  
 
Set to True, to use ClassifyTest method instead of the method assigned to OnClassifyTest.  
 
Called by LearnData method and K-NN to request the positioning to the first record.  
 
Called by LearnData method and K-NN when the last record is fetched.  
 
If the response of all classes is below RejectProbability the example will not be classified to any of the known classes.  
 
Name 
Description 
 
Called by PrePrune, PostPrune and Prune methods and ClassifyTest methods.  
 
Name 
Description 
 
Total number of attributes.  
 
Total number of discrete valued attributes or fields holding discreting values in the dataset.  
 
Defines the distance model used, when calculating the distance between examples.  
 
Total number of learned examples.  
 
Total number of real valued attributes or fields holding real values in the dataset.  
 
Set to True, if Class indexes are zero based.  
 
Defines the K parameter for the K-NN algorithm.  
 
Set the value of this property to the index of the example that you want to be ignored during the classification.  
 
Specifies the value indicating a "missing value" (no entry) for discrete attributes in the dataset.  
 
Specifies the value indicating a "missing value" (no entry) for real valued attributes in the dataset.  
 
If true, the number of attributes compared will be normalized between comparisons.  
 
Set this property to true, to store all learned examples.  
Copyright (c) 1999-2017 by Dew Research. All rights reserved.
What do you think about this topic? Send feedback!