%iKnow.Classification.Methods.VectorSpace
abstract class %iKnow.Classification.Methods.VectorSpace extends %iKnow.Classification.Methods.Base
This Builder Method implementation generates category term weights as a vector per category that
exists in the same vector space as the document term vector. The similarity between the
document vector and each of the category vectors can then be calculated using euclidean distance
or cosine similarity (angle). Alternatively, these weights can be used for a linear regression
formula, calculating a score rather than a similarity/distance.
A combination of global term weights (across the corpus), local term weights (within each
category) and normalization (per category) is used to calculate these category vectors.
Property Inventory
Method Inventory
PropertiesCopy link to this section
property CategoryGlobalTermWeights as %String (VALUELIST = ",none,IDF") [ InitialExpression = "IDF" ];
Copy link to this section
The corpus-wide relevancy factor to take into account when calculating term weights
Property methods: CategoryGlobalTermWeightsDisplayToLogical(), CategoryGlobalTermWeightsGet(), CategoryGlobalTermWeightsIsValid(), CategoryGlobalTermWeightsLogicalToDisplay(), CategoryGlobalTermWeightsLogicalToOdbc(), CategoryGlobalTermWeightsNormalize(), CategoryGlobalTermWeightsSet()
property CategoryLocalTermMetric as %String (VALUELIST = ",frequency,spread") [ InitialExpression = "spread" ];
Copy link to this section
The metric to use for calculating the local term weights
Property methods: CategoryLocalTermMetricDisplayToLogical(), CategoryLocalTermMetricGet(), CategoryLocalTermMetricIsValid(), CategoryLocalTermMetricLogicalToDisplay(), CategoryLocalTermMetricLogicalToOdbc(), CategoryLocalTermMetricNormalize(), CategoryLocalTermMetricSet()
property CategoryLocalTermWeights as %String (VALUELIST = ",binary,linear,logarithmic") [ InitialExpression = "linear" ];
Copy link to this section
The per-category term weight factor
Property methods: CategoryLocalTermWeightsDisplayToLogical(), CategoryLocalTermWeightsGet(), CategoryLocalTermWeightsIsValid(), CategoryLocalTermWeightsLogicalToDisplay(), CategoryLocalTermWeightsLogicalToOdbc(), CategoryLocalTermWeightsNormalize(), CategoryLocalTermWeightsSet()
property CategoryNormalization as %String (VALUELIST = ",none,cosine") [ InitialExpression = "none" ];
Copy link to this section
Whether and how to normalize the category vectors
Property methods: CategoryNormalizationDisplayToLogical(), CategoryNormalizationGet(), CategoryNormalizationIsValid(), CategoryNormalizationLogicalToDisplay(), CategoryNormalizationLogicalToOdbc(), CategoryNormalizationNormalize(), CategoryNormalizationSet()
property CustomTermWeights [ MultiDimensional ];
Copy link to this section
Property methods: CustomTermWeightsDisplayToLogical(), CustomTermWeightsGet(), CustomTermWeightsIsValid(), CustomTermWeightsLogicalToDisplay(), CustomTermWeightsLogicalToOdbc(), CustomTermWeightsNormalize(), CustomTermWeightsSet()
MethodsCopy link to this section
method %OnRemoveTerm(pIndex As %Integer ) as %Status
Copy link to this section
Callback invoked whenever an entire term at an index has been removed
Sets a custom weight factor for the term at pIndex in pCategory .
method %SetCustomWeights(pIndex As %Integer , ByRef pCustomWeights) as %Status
Copy link to this section
Inherited MembersCopy link to this section
Inherited Properties
Inherited Methods
SubclassesCopy link to this section