# Linguistics in Vespa

Vespa uses a linguistics module to process text in queries and documents during indexing and searching. The goal of linguistic processing is to increase recall (how many documents are matched) without hurting precision (the relevance of the documents matched) too much. It consists of such operations as:

• tokenizing text into chunks of known types such as words and punctuation.
• normalizing accents.
• finding the base form of words (stemming or lemmatization).

Linguistic processing is run when writing documents, and when querying:

The processing is run on string fields with index indexing mode. Overview:

1. When writing documents, string fields with indexing: index are by default processed. A field's language will configure this processing. A document/fields can have the language set explicitly, if not, it is detected.
2. The field's content is tokenized, normalized and stemmed, and the resulting terms are added to the index.
3. A query is also processed the same way - this is symmetric with the schema. The language of the strings is detected unless specified using model.locale or annotations like language.
4. The processed query is evaluated on the content nodes, and will only work as expected if both documents and queries are processed using identical settings (like same language).

These operations can be turned on or off per field in the schema. See implicitTransforms for how to enable/disable transforms per query term.

The default linguistics module is OpenNlp.

See Text Matching and Ranking for examples and how to experiment with data.

## Creating a custom linguistics implementation

A linguistics component is an implementation of com.yahoo.language.Linguistics. Refer to the com.yahoo.language.simple.SimpleLinguistics implementation (which can be subclassed for convenience).

SimpleLinguistics provides support for english stemming only. Try loading the com.yahoo.language.simple.SimpleLinguistics module, or providing another linguistics module.

The linguistics implementation must be configured as a component in container clusters doing linguistics processing, see injecting components.

As document processing for indexing is by default done by an autogenerated container cluster which cannot be configured, specify a container cluster for indexing explicitly.

This example shows how to configure SimpleLinguistics for linguistics using the same cluster for both query and indexing processing (if using different clusters, add the same linguistics component to all of them):

<services>

<container version="1.0" id="mycontainer">
<component id="com.yahoo.language.simple.SimpleLinguistics"/>
<document-processing/>
<search/>
<nodes ...>
</container>

<content version="1.0">
<redundancy>1</redundancy>
<documents>
<document type="mydocument" mode="index"/>
<document-processing cluster="mycontainer"/>
</documents>
<nodes ...>
</content>

</services>


If changing the linguistics component of a live system, recall can be reduced until all documents are re-written. This because documents will still be stored with tokens generated by the previous linguistics module.

## Language handling

Vespa does not know the language of a document - this applies:

1. The indexing processor is instructed on a per-field level what language to use when calling the underlying linguistics library
2. The query processor is instructed on a per-query level what language to use

If no language is explicitly set in a document or a query, Vespa will run its configured language detector on the available text (the full content of a document field, or the full query= parameter value).

A document that contains the exact same word as a query might not be recall-able if the language of the document field is detected differently from the query. Unless the query has explicitly declared a language, this can occur.

### Indexing with language

The indexing process run by Vespa is a sequential execution of the indexing scripts of each field in the schema, in the declared order. At any point, the script may set the language that will be used for indexing statements for subsequent fields, using set_language. Example:

schema doc {
document doc {
field language type string {
indexing: set_language
}
field title type string {
indexing: index
}
}
}


If a language has not been set when tokenization of a field is run, the language is determined by language detection.

If all documents have the same language, the language can be hardcoded it the schema in this way:

schema doc {

field language type string {
indexing: "en" | set_language
}

document doc {
...


If the same document contains fields in multiple languages, set_language can be invoked multiple times, e.g

schema doc {
document doc {
field language_title1 type string {
indexing: set_language
}
field title1 type string {
indexing: index
}
field language_title2 type string {
indexing: set_language
}
field title2 type string {
indexing: index
}
}
}


Or, if fixed per field, use multiple indexing statements in each field:

schema doc {
document doc {
field my_english_field type string {
indexing {
"en" | set_language;
index;
}
}
field my_spanish_field type string {
indexing {
"es" | set_language;
index;
}
}
}
}


### Field language detection

When indexing a document, if a field has unknown language (i.e. not set using set_language), language detection is run on the field's content. This means, language detection is per field, not per document.

See query language detection for detection confidence, fields with little text will default to English.

### Querying with language

The content of an indexed string field is language-agnostic. One must therefore apply a symmetric tokenization on the query terms in order to match the content of that field.

The query parser subscribes to configuration that tells it what fields are indexed strings, and every query term that targets such a field are run through appropriate tokenization. The language query parameter controls the language state of these calls.

Because an index may simultaneously contain terms in any number of languages, one can have stemmed variants of one language match the stemmed variants of another. To work around this, store the language of a document in a separate attribute, and apply a filter against that attribute at query-time.

By default, there is no knowledge anywhere that captures what languages are used to generate the content of an index. The language parameter only affects the transformation of query terms that hit tokenized indexes.

### Query language detection

If no language parameter is used, or the query terms are annotated, the language detector is called to process the query string.

Queries are normally short, as a consequence, the detection confidence is low. Example:

$vespa query "select * from music where userInput(@text)" \ tracelevel=3 text='Eine kleine Nachtmusik' | grep 'Stemming with language' "message": "Stemming with language=ENGLISH"$ vespa query "select * from music where userInput(@text)" \
tracelevel=3 text='Eine kleine Nachtmusik schnell' | grep 'Stemming with language'
"message": "Stemming with language=GERMAN"


See #24265 for details - in short, with the current 0.02 confidence cutoff, queries with 3 terms or less will default to English.

### Multiple languages

Approaches:

1. Use semantic retrieval using a sentence encoder, which has been trained on multi-lingual corpus like the one used in Semantic QA retrieval. Sentence-Bert is another. First phase matching is then done using approximate nearest neighbor: #9747. Query terms can also be passed down for fine-tuning of precision.
2. Stem and tokenize the query using the relevant languages, build a query tree using weakAnd / or and using equiv per stem variant. This is easiest done in a custom Searcher as mentioned in #12154.

Example:

lang=fr: machine learning => machin learn

lang=en: machine learning => machine learn

Using weakAnd here as example as that technique is already mentioned in #12154:

select * from sources * where rank(weakAnd(default contains equiv("machin", "machine"), default contains "learn)), default contains "machine", default contains "learning")


We now retrieve using all possible stems/base forms with weakAnd, and use the rank operator to pass in the original query form, so that ranking can rank literal matches (original) higher. Benefit of equiv is that it allows multiple term variants to share the same position, so that proximity ranking does not become broken by this approach.

## Tokenization

Tokenization removes any non-word characters, and splits the string into tokens on each word boundary. In addition, CJK tokens are split using a segmentation algorithm. The resulting tokens are then searchable in the index.

To index strings as-is (that is, avoid tokenization), use indexing-rewrite: none.

Also see N-gram matching.

## Normalization

An example normalization is à ⇒ a. Normalizing will cause accents and similar decorations which are often misspelled to be normalized the same way both in documents and queries.

Vespa uses java.text.Normalizer to normalize text, see SimpleTransformer.java. Normalization preserves case.

Refer to the nfkc query term annotation. Also see the YQL accentDrop annotation.

## Stemming

Stemming means translate a word to its base form (singular forms for nouns, infinitive for verbs), using a stemmer. Use of stemming increases search recall, because the searcher is usually interested in documents containing query words regardless of the word form used. Stemming in Vespa is symmetric, i.e. words are converted to stems both when indexing and searching.

Examples of this is when text is indexed, the stemmer will convert the noun reports (plural) to report, and the latter will be stored in the index. Likewise, before searching, reports will be stemmed to report. Another example is that am, are and was will be stemmed to be both in queries and indexes.

When bolding is enabled, all forms of the query term will be bolded. I.e. when searching for reports, both report, reported and reports will be bolded.

See the stem query term annotation.

### Theory

From a matching point of view, stemming takes all possible token strings and maps them into equivalence classes. So in the example above, the set of tokens { report, reports, reported } are in an equivalence class. To represent the class, the linguistics library should pick the best element in the class. At query time, the text typed by a user will be tokenized, and then each token should be mapped to the most likely equivalence class, again represented by the shortest element that belongs to the class.

While the theory sounds pretty simple, in practice it is not always possible to figure out which equivalence class a token should belong to. A typical example is the string number. In most cases we would guess this to mean a numerical entity of some kind, and the equivalence class would be { number, numbers } - but it could also be a verb, with a different equivalence class { number, numbered, numbering }. These are of course closely related, and in practice they will be merged, so we'll have a slightly larger equivalence class { number, numbers, numbered, numbering } and be happy with that. However, in a sentence such as my legs keep getting number every day, the number token clearly does not have the semantics of a numerical entity, but should be in the equivalence class { numb, number, numbest, numbness } instead. But blindly assigning number to the equivalence class numb is clearly not right, since the more numb meaning is much less likely than the numerical entity meaning.

The approach currently taken by the low-level linguistics library will often lead to problems in the number-like cases as described above. To give better recall, Vespa has implemented a multiple stemming option.

### Configuration

By default, all words are stemmed to their best form. Refer to the stemming reference for other stemming types. To change type, add:

stemming: [stemming-type]


Stemming can be set either for a field, a fieldset or as a default for all fields. Example: Disable stemming for the field title:

field title type string {
indexing: summary | index
stemming: none
}


See andSegmenting for how to control re-segmenting when stemming.