What is QueryProcessor-OpenSearch(Open Search)-阿里云帮助中心

A QueryProcessor defines a query analysis rule for an OpenSearch application. It specifies which processors to apply, which indexes the rule targets, and whether the rule is the active default.

Structure

{
    "name": "qp_lsh_test_1",
    "domain": "GENERAL",
    "processors": [{
        "name": "stop_word",
        "use_system_dictionary": true,
        "intervention_dictionary": ""
    }],
    "indexes": [
        "default"
    ],
    "active": true
}

Field	Type	Description
`name`	String	The name of the query analysis rule. Can contain lowercase letters, digits, and underscores (`_`). Must start with a lowercase letter. Maximum 16 characters.
`domain`	String	The industry domain the rule applies to. Valid values: `GENERAL`, `ECOMMERCE`, `IT_CONTENT`.
`processors[]`	Object	The processors to apply in this rule. Each entry is a Processor object. See Processor below.
`indexes`	Array	The indexes this rule applies to.
`active`	Boolean	Whether this rule is the default query analysis rule.

Processor

A processor defines a single query analysis feature and its configuration.

OpenSearch supports six processor types:

Processor name	Feature
`stop_word`	Stop word filtering
`spell_check`	Spelling correction
`term_weighting`	Term weight analysis
`synonym`	Synonym configuration
`category_prediction`	Category prediction — see Category prediction
`ner`	Named Entity Recognition (NER) — see NER

Fields

{
    "name": "stop_word",
    "use_system_dictionary": true,
    "intervention_dictionary": ""
}

Field	Type	Description
`name`	String	The processor type. Valid values: `stop_word`, `spell_check`, `term_weighting`, `synonym`, `category_prediction`, `ner`.
`use_system_dictionary`	Boolean	Specifies whether to use built-in dictionaries.
`intervention_dictionary`	String	The name of the intervention dictionary. Leave blank if not applicable.

Category prediction

Use the category_prediction processor to apply a trained category prediction model.

Fields

{
    "name": "stop_word",
    "categoryPrediction": 12345,
    "projectId": 12346
}

Field	Type	Description
`name`	String	Must be `category_prediction`.
`categoryPrediction`	Integer	The ID of the category prediction model.
`projectId`	Integer	The ID of the algorithm project.

NER

Use the ner processor to apply Named Entity Recognition (NER) and control the priority of recognized entity types.

Fields

{
    "name": "stop_word",
    "use_system_dictionary": true,
    "intervention_dictionary": "",
    "priorities": [{
        "priority": "HIGH",
        "tag": "test",
        "order": 1
    }]
}

Field	Type	Description
`name`	String	Must be `ner`.
`use_system_dictionary`	Boolean	Specifies whether to use built-in dictionaries.
`intervention_dictionary`	String	The name of the intervention dictionary. Leave blank if not applicable.
`priorities[]`	Object	Priority settings for entity types.
`priorities[].priority`	String	The priority level. Valid values: `HIGH`, `MIDDLE`, `LOW`.
`priorities[].tag`	String	The internal name of the entity type.
`priorities[].order`	Integer	The rank within entity types that share the same priority level. A lower value means higher priority. Default: `0`.