Class PassageScorer

Ranks passages found by ICUPostingsHighlighter.

Each passage is scored as a miniature document within the document. The final score is computed as norm * ∑ (weight * tf). The default implementation is norm * BM25.

This is a Lucene.NET EXPERIMENTAL API, use at your own risk
Inheritance
System.Object
PassageScorer
Namespace: Lucene.Net.Search.PostingsHighlight
Assembly: Lucene.Net.ICU.dll
Syntax
public class PassageScorer : object

Constructors

| Improve this Doc View Source

PassageScorer()

Creates PassageScorer with these default values:

  • k1 = 1.2
  • b = 0.75
  • pivot = 87

Declaration
public PassageScorer()
| Improve this Doc View Source

PassageScorer(Single, Single, Single)

Creates PassageScorer with specified scoring parameters

Declaration
public PassageScorer(float k1, float b, float pivot)
Parameters
Type Name Description
System.Single k1

Controls non-linear term frequency normalization (saturation).
System.Single b

Controls to what degree passage length normalizes tf values.
System.Single pivot

Pivot value for length normalization (some rough idea of average sentence length in characters).

Methods

| Improve this Doc View Source

Norm(Int32)

Normalize a passage according to its position in the document.

Typically passages towards the beginning of the document are more useful for summarizing the contents.

The default implementation is 1 + 1/log(pivot + passageStart)

Declaration
public virtual float Norm(int passageStart)
Parameters
Type Name Description
System.Int32 passageStart

start offset of the passage
Returns
Type Description
System.Single

a boost value multiplied into the passage's core.
| Improve this Doc View Source

Tf(Int32, Int32)

Computes term weight, given the frequency within the passage and the passage's length.

Declaration
public virtual float Tf(int freq, int passageLen)
Parameters
Type Name Description
System.Int32 freq

number of occurrences of within this passage
System.Int32 passageLen

length of the passage in characters.
Returns
Type Description
System.Single

term weight
| Improve this Doc View Source

Weight(Int32, Int32)

Computes term importance, given its in-document statistics.

Declaration
public virtual float Weight(int contentLength, int totalTermFreq)
Parameters
Type Name Description
System.Int32 contentLength

length of document in characters
System.Int32 totalTermFreq

number of time term occurs in document
Returns
Type Description
System.Single

term importance