public abstract class Similarity extends Object implements Serializable
Similarity defines the components of Lucene scoring. Overriding computation of these components is a convenient way to alter Lucene scoring.
Suggested reading: Introduction To Information Retrieval, Chapter 6.
The following describes how Lucene scoring evolves from underlying information retrieval models to (efficient) implementation. We first brief on VSM Score, then derive from it Lucene's Conceptual Scoring Formula, from which, finally, evolves Lucene's Practical Scoring Function (the latter is connected directly with Lucene classes and methods).
Lucene combines Boolean model (BM) of Information Retrieval with Vector Space Model (VSM) of Information Retrieval - documents "approved" by BM are scored by VSM.
In VSM, documents and queries are represented as weighted vectors in a multi-dimensional space, where each distinct index term is a dimension, and weights are Tf-idf values.
VSM does not require weights to be Tf-idf values, but Tf-idf values are believed to produce search results of high quality, and so Lucene is using Tf-idf. Tf and Idf are described in more detail below, but for now, for completion, let's just say that for given term t and document (or query) x, Tf(t,x) varies with the number of occurrences of term t in x (when one increases so does the other) and idf(t) similarly varies with the inverse of the number of index documents containing term t.
VSM score of document d for query q is the
Cosine Similarity
of the weighted query vectors V(q) and V(d):
| ||||||
|
Note: the above equation can be viewed as the dot product of the normalized weighted vectors, in the sense that dividing V(q) by its euclidean norm is normalizing it to a unit vector.
Lucene refines VSM score for both search quality and usability:
Under the simplifying assumption of a single field in the index,
we get Lucene's Conceptual scoring formula:
| |||||||
|
The conceptual formula is a simplification in the sense that (1) terms and documents are fielded and (2) boosts are usually per query term rather than per query.
We now describe how Lucene implements this conceptual scoring formula, and derive from it Lucene's Practical Scoring Function.
For efficient score computation some scoring components are computed and aggregated in advance:
Lucene's Practical Scoring Function is derived from the above. The color codes demonstrate how it relates to those of the conceptual formula:
| |||||||
|
where
DefaultSimilarity
is:
tf(t in d) =
|
frequency½ |
DefaultSimilarity
is:
idf(t) =
|
1 + log ( |
|
) |
coord(q,d)
by the Similarity in effect at search time.
DefaultSimilarity
produces a Euclidean norm:
queryNorm(q) =
queryNorm(sumOfSquaredWeights)
=
|
|
Weight
object.
For example, a BooleanQuery
computes this value as:
sumOfSquaredWeights =
q.getBoost() 2
·
|
∑ | ( idf(t) · t.getBoost() ) 2 |
t in q |
setBoost()
.
Notice that there is really no direct API for accessing a boost of one term in a multi term query,
but rather multi terms are represented in a query as multi
TermQuery
objects,
and so the boost of a term in the query is accessible by calling the sub-query
getBoost()
.
doc.setBoost()
before adding the document to the index.
field.setBoost()
before adding the field to a document.
computeNorm(java.lang.String, org.apache.lucene.index.FieldInvertState)
method is responsible for
combining all of these factors into a single float.
When a document is added to the index, all the above factors are multiplied.
If the document has multiple fields with the same name, all their boosts are multiplied together:
norm(t,d) =
doc.getBoost()
·
lengthNorm
·
|
∏ |
f.getBoost ()
|
field f in d named as t |
encoded
as a single byte
before being stored.
At search time, the norm byte value is read from the index
directory
and
decoded
back to a float norm value.
This encoding/decoding, while reducing index size, comes with the price of
precision loss - it is not guaranteed that decode(encode(x)) = x.
For instance, decode(encode(0.89)) = 0.75.
Similarity
for search.
Modifier and Type | Field and Description |
---|---|
static int |
NO_DOC_ID_PROVIDED |
Constructor and Description |
---|
Similarity() |
Modifier and Type | Method and Description |
---|---|
abstract float |
computeNorm(String field,
FieldInvertState state)
Computes the normalization value for a field, given the accumulated
state of term processing for this field (see
FieldInvertState ). |
abstract float |
coord(int overlap,
int maxOverlap)
Computes a score factor based on the fraction of all query terms that a
document contains.
|
static float |
decodeNorm(byte b)
Deprecated.
Use
decodeNormValue(byte) instead. |
float |
decodeNormValue(byte b)
Decodes a normalization factor stored in an index.
|
static byte |
encodeNorm(float f)
Deprecated.
Use
encodeNormValue(float) instead. |
byte |
encodeNormValue(float f)
Encodes a normalization factor for storage in an index.
|
static Similarity |
getDefault()
Return the default Similarity implementation used by indexing and search
code.
|
static float[] |
getNormDecoder()
Deprecated.
Use instance methods for encoding/decoding norm values to enable customization.
|
abstract float |
idf(int docFreq,
int numDocs)
Computes a score factor based on a term's document frequency (the number
of documents which contain the term).
|
Explanation.IDFExplanation |
idfExplain(Collection<Term> terms,
Searcher searcher)
Computes a score factor for a phrase.
|
Explanation.IDFExplanation |
idfExplain(Term term,
Searcher searcher)
This method forwards to
idfExplain(Term,Searcher,int) by passing
searcher.docFreq(term) as the docFreq. |
Explanation.IDFExplanation |
idfExplain(Term term,
Searcher searcher,
int docFreq)
Computes a score factor for a simple term and returns an explanation
for that score factor.
|
float |
lengthNorm(String fieldName,
int numTokens)
Deprecated.
Please override computeNorm instead
|
abstract float |
queryNorm(float sumOfSquaredWeights)
Computes the normalization value for a query given the sum of the squared
weights of each of the query terms.
|
float |
scorePayload(int docId,
String fieldName,
int start,
int end,
byte[] payload,
int offset,
int length)
Calculate a scoring factor based on the data in the payload.
|
static void |
setDefault(Similarity similarity)
Set the default Similarity implementation used by indexing and search
code.
|
abstract float |
sloppyFreq(int distance)
Computes the amount of a sloppy phrase match, based on an edit distance.
|
abstract float |
tf(float freq)
Computes a score factor based on a term or phrase's frequency in a
document.
|
float |
tf(int freq)
Computes a score factor based on a term or phrase's frequency in a
document.
|
public static final int NO_DOC_ID_PROVIDED
public static void setDefault(Similarity similarity)
public static Similarity getDefault()
This is initially an instance of DefaultSimilarity
.
@Deprecated public static float decodeNorm(byte b)
decodeNormValue(byte)
instead.decodeNormValue(byte)
public float decodeNormValue(byte b)
WARNING: If you override this method, you should change the default
Similarity to your implementation with setDefault(Similarity)
.
Otherwise, your method may not always be called, especially if you omit norms
for some fields.
encodeNormValue(float)
@Deprecated public static float[] getNormDecoder()
encodeNormValue(float)
,
decodeNormValue(byte)
public abstract float computeNorm(String field, FieldInvertState state)
FieldInvertState
).
Implementations should calculate a float value based on the field state and then return that value.
Matches in longer fields are less precise, so implementations of this
method usually return smaller values when state.getLength()
is large,
and larger values when state.getLength()
is small.
Note that the return values are computed under
IndexWriter.addDocument(org.apache.lucene.document.Document)
and then stored using
encodeNormValue(float)
.
Thus they have limited precision, and documents
must be re-indexed if this method is altered.
For backward compatibility this method by default calls
lengthNorm(String, int)
passing
FieldInvertState.getLength()
as the second argument, and
then multiplies this value by FieldInvertState.getBoost()
.
field
- field namestate
- current processing state for this field@Deprecated public final float lengthNorm(String fieldName, int numTokens)
Matches in longer fields are less precise, so implementations of this
method usually return smaller values when numTokens
is large,
and larger values when numTokens
is small.
Note that the return values are computed under
IndexWriter.addDocument(org.apache.lucene.document.Document)
and then stored using
encodeNormValue(float)
.
Thus they have limited precision, and documents
must be re-indexed if this method is altered.
fieldName
- the name of the fieldnumTokens
- the total number of tokens contained in fields named
fieldName of doc.AbstractField.setBoost(float)
public abstract float queryNorm(float sumOfSquaredWeights)
This does not affect ranking, but the default implementation does make scores from different queries more comparable than they would be by eliminating the magnitude of the Query vector as a factor in the score.
sumOfSquaredWeights
- the sum of the squares of query term weightspublic byte encodeNormValue(float f)
The encoding uses a three-bit mantissa, a five-bit exponent, and the zero-exponent point at 15, thus representing values from around 7x10^9 to 2x10^-9 with about one significant decimal digit of accuracy. Zero is also represented. Negative numbers are rounded up to zero. Values too large to represent are rounded down to the largest representable value. Positive values too small to represent are rounded up to the smallest positive representable value.
WARNING: If you override this method, you should change the default
Similarity to your implementation with setDefault(Similarity)
.
Otherwise, your method may not always be called, especially if you omit norms
for some fields.
AbstractField.setBoost(float)
,
SmallFloat
@Deprecated public static byte encodeNorm(float f)
encodeNormValue(float)
instead.f
- norm-value to encodeencodeNormValue(float)
public float tf(int freq)
idf(int, int)
factor for each term in the query and these products are then summed to
form the initial score for a document.
Terms and phrases repeated in a document indicate the topic of the
document, so implementations of this method usually return larger values
when freq
is large, and smaller values when freq
is small.
The default implementation calls tf(float)
.
freq
- the frequency of a term within a documentpublic abstract float sloppyFreq(int distance)
tf(float)
.
A phrase match with a small edit distance to a document passage more closely matches the document, so implementations of this method usually return larger values when the edit distance is small and smaller values when it is large.
distance
- the edit distance of this sloppy phrase matchPhraseQuery.setSlop(int)
public abstract float tf(float freq)
idf(int, int)
factor for each term in the query and these products are then summed to
form the initial score for a document.
Terms and phrases repeated in a document indicate the topic of the
document, so implementations of this method usually return larger values
when freq
is large, and smaller values when freq
is small.
freq
- the frequency of a term within a documentpublic Explanation.IDFExplanation idfExplain(Term term, Searcher searcher, int docFreq) throws IOException
The default implementation uses:
idf(searcher.docFreq(term), searcher.maxDoc());Note that
Searcher.maxDoc()
is used instead of
IndexReader#numDocs()
because also
Searcher.docFreq(Term)
is used, and when the latter
is inaccurate, so is Searcher.maxDoc()
, and in the same direction.
In addition, Searcher.maxDoc()
is more efficient to computeterm
- the term in questionsearcher
- the document collection being searchedIOException
public Explanation.IDFExplanation idfExplain(Term term, Searcher searcher) throws IOException
idfExplain(Term,Searcher,int)
by passing
searcher.docFreq(term)
as the docFreq.
WARNING: if you subclass Similariary and override this
method then you may hit a peformance hit for certain
queries. Better to override idfExplain(Term,Searcher,int)
instead.IOException
public Explanation.IDFExplanation idfExplain(Collection<Term> terms, Searcher searcher) throws IOException
The default implementation sums the idf factor for each term in the phrase.
terms
- the terms in the phrasesearcher
- the document collection being searchedIOException
public abstract float idf(int docFreq, int numDocs)
tf(int)
factor for each term in the query and these products are
then summed to form the initial score for a document.
Terms that occur in fewer documents are better indicators of topic, so implementations of this method usually return larger values for rare terms, and smaller values for common terms.
docFreq
- the number of documents which contain the termnumDocs
- the total number of documents in the collectionpublic abstract float coord(int overlap, int maxOverlap)
The presence of a large portion of the query terms indicates a better match with the query, so implementations of this method usually return larger values when the ratio between these parameters is large and smaller values when the ratio between them is small.
overlap
- the number of query terms matched in the documentmaxOverlap
- the total number of terms in the querypublic float scorePayload(int docId, String fieldName, int start, int end, byte[] payload, int offset, int length)
The default implementation returns 1.
docId
- The docId currently being scored. If this value is NO_DOC_ID_PROVIDED
, then it should be assumed that the PayloadQuery implementation does not provide document informationfieldName
- The fieldName of the term this payload belongs tostart
- The start position of the payloadend
- The end position of the payloadpayload
- The payload byte array to be scoredoffset
- The offset into the payload arraylength
- The length in the array