Product Selector

Fusion 5.9
    Fusion 5.9

    Recommend Similar QueriesQuery pipeline stage configuration specifications

    Table of Contents

    This stage provides query-to-query recommendations.

    The stage works in the following way:

    1. Perform an exact match of the incoming query to queries stored in the similar queries collection. Return results if a match is found.

    2. If Match non-exact matches is enabled, the pipeline attempts a normal query and returns the results.

    3. If Enable POS tagging is checked, the stage will perform POS tagging on the incoming query and use only the noun parts for the non-exact match.

    The spaCy model for part-of-speech (POS) tagging no longer ships with fusion. It needs to be deployed as a Seldon container using the Create Seldon Model deployment job.

    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.

    This stage returns query recommendations

    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.

    collection - stringrequired

    Collection that contains job results from similar queries spark job

    nonExactMatches - boolean

    If phrase search does not return any results, search only using nouns (annotated by open-nlp) model. Turn this off to force for exact matches

    Default: true

    modelId - string

    Model ID. Model id available in ml-model-service. Default models are opennlp and spacy

    Default: opennlp

    posTagging - boolean

    Enable POS Tagging to tag nouns in the query and search with the found nouns

    Default: true