Product Selector

Fusion 5.12
    Fusion 5.12

    Text Tagger Stage

    Table of Contents

    This stage uses the SolrTextTagger handler to identify known entities in the query by searching either of the following:

    To perform query rewriting, this stage searches for matching instances of:

    When the query rewrite entails boosting, the boosting is applied later in the pipeline, during the Solr Query stage.

    The below diagram shows the process flow for the Text Tagger Stage:

    Text Tagger Stage Process

    The underlying SolrTextTagger currently only supports single-shard collections. Fusion users should ensure their _query_rewrite collection, or whatever collection the Text Tagger stage is configured to use, is single-sharded before enabling this stage.
    Although this stage is available without a Fusion AI license, it is only effective after running Fusion AI jobs or creating Fusion AI rules. See Query Rewriting for details.

    Configuration

    When entering configuration values in the UI, use unescaped characters, such as \t for the tab character. When entering configuration values in the API, use escaped characters, such as \\t for the tab character.

    Queries a Solr text tagger request handler to perform spell correction, phrase boosting, and synonym expansion.

    skip - boolean

    Set to true to skip this stage.

    Default: false

    label - string

    A unique label for this stage.

    <= 255 characters

    condition - string

    Define a conditional script that must result in true or false. This can be used to determine if the stage should process or not.

    taggerCollectionId - string

    Collection to send the tagger request to; defaults to the query_rewrite collection for the application in context. Collection must contain only one shard, as the underlying Text Tagger in Solr doesn't currently support multi-shard collections.

    >= 1 characters

    paramToTag - string

    Name of the parameter in the request containing text to tag, defaults to 'q'

    Default: q

    taggedOutputParam - string

    Apply the matching tags to the 'paramToTag' value and set the parameter specified by this option; defaults to the value of the 'paramToTag' setting.

    saveTagsInContextKey - string

    Save tags in context instead of applying directly to the incoming query in this stage; allows downstream stages to apply the tags after doing other processing.

    spell_corrections_enabled - boolean

    If checked, then this stage will perform spell corrections on the incoming query.

    Default: true

    phrase_boosting_enabled - boolean

    If checked, then this stage will perform phrase boosting on the incoming query.

    Default: true

    synonym_expansion_enabled - boolean

    If checked, then this stage will perform synonym expansion on the incoming query.

    Default: true

    tail_rewrites_enabled - boolean

    If checked, then this stage will perform tail rewrites on the incoming query.

    Default: true

    filterOverride - string

    Use this option to override filtering for built-in tagger doc types with your own filter.

    synonymExpansionBoost - number

    Boost applied to the original term when doing synonym expansion; set to -1 to disable this behavior.

    Default: 2

    phraseBoost - number

    Default boost to be applied to phrases that don't have a boost set; set to -1 to disable this behavior.

    Default: 2

    phraseSlop - integer

    Default phrase slop to be applied to detected phrases.

    Default: 10

    overlaps - string

    Choose the algorithm to determine which tags in an overlapping set should be retained, versus being pruned away. Options are: ALL, NO_SUB, LONGEST_DOMINANT_RIGHT; defaults to ALL

    Default: ALL

    Allowed values: ALLNO_SUBLONGEST_DOMINANT_RIGHT

    params - array[object]

    object attributes:{key required : {
     display name: Parameter Name
     type: string
    }
    value : {
     display name: Parameter Value
     type: string
    }
    }

    useEmbeddedSolrServer - boolean

    If checked, then this stage will load the remote tagger collection into an embedded Solr to perform the lookup in memory. Only use this option if the tagger collection fits in the memory available to the Fusion API service.

    Default: false