Class DFRSimilarity
Implements the divergence from randomness (DFR) framework introduced in Gianni Amati and Cornelis Joost Van Rijsbergen. 2002. Probabilistic models of information retrieval based on measuring the divergence from randomness. ACM Trans. Inf. Syst. 20, 4 (October 2002), 357-389.
The DFR scoring formula is composed of three separate components: the basic model, the aftereffect and an additional normalization component, represented by the classes BasicModel, AfterEffect and Normalization, respectively. The names of these classes were chosen to match the names of their counterparts in the Terrier IR engine.
To construct a DFRSimilarity, you must specify the implementations for all three components of DFR:
| ComponentImplementations | |
|---|---|
BasicModel: Basic model of information content:
| |
AfterEffect: First normalization of information gain:
| |
Normalization: Second (length) normalization:
|
Note that qtf, the multiplicity of term-occurrence in the query, is not handled by this implementation.
@lucene.experimentalInherited Members
Assembly: DistributedLucene.Net.dll
Syntax
public class DFRSimilarity : SimilarityBaseConstructors
| Name | Description |
|---|---|
| DFRSimilarity(BasicModel, AfterEffect, Normalization) | Creates DFRSimilarity from the three components.
Note that |
Fields
| Name | Description |
|---|---|
| m_afterEffect | The first normalization of the information content. |
| m_basicModel | The basic model for information content. |
| m_normalization | The term frequency normalization. |
Properties
| Name | Description |
|---|---|
| AfterEffect | Returns the first normalization |
| BasicModel | Returns the basic model of information content |
| Normalization | Returns the second normalization |
Methods
| Name | Description |
|---|---|
| Explain(Explanation, BasicStats, Int32, Single, Single) | |
| Score(BasicStats, Single, Single) | |
| ToString() |
