Filter
A filter to use to restrict the resources to return.
You specify a filter as a percent-encoded JSON object that describes the filter to apply. The following table describes the available properties in the filter JSON object, and the values of the Type parameter that you can use the filter for.
Property | Type values | Description | Format |
---|---|---|---|
text
|
question
question_equivalence_class |
A text filter that matches the text in questions and rules in the items that GetResources returns. You can use any IDOL query text syntax, such as keywords, Boolean and proximity operators, and Wildcard characters. |
{ "text": "some search text" }
|
conceptual
|
question_equivalence_class | A Boolean value that specifies whether to use conceptual matching for the specified text . This option has an effect only when you use it with a text filter. Set conceptual to true to match the query text against all the questions in your Answer Bank, and return the parent question equivalence class for matching questions. This option might provide better matches when searching for equivalent questions.You can also use conceptual_minscore to set a matching threshold. |
{ "text": "some search text", "conceptual": true |
conceptual_minscore
|
question_equivalence_class | The minimum percentage score that a question match must have for GetResources to return the question equivalence class. This option has an effect only when you use it with a text filter with conceptual set to true.The default value is 10. |
{ "text": "some search text", "conceptual": true, "conceptual_minscore": 20 |
answer_text
|
question_equivalence_class | A text filter that matches the answer text in the question equivalence classes that GetResources returns. You can use any IDOL query text syntax, such as keywords, Boolean and proximity operators, and Wildcard characters. | { "answer_text": "some search text" }
|
reference_question_text
|
question_equivalence_class | A text filter that matches the reference question text in the question equivalence classes that GetResources returns. You can use any IDOL query text syntax, such as keywords, Boolean and proximity operators, and Wildcard characters. | { "reference_question_text": "some search text" }
|
rule_suggestion_text
|
rule_suggestion | An array of additional questions to include in the rule suggestion. You can use this option to suggest a rule for a set of questions that do not exist in the Answer Bank index. | { "rule_suggestion_text": ["some search text"] }
|
state
|
question
question_equivalence_class |
A question or question equivalence class state filter that allows you to find items with a particular state. The possible question state values are: incoming , answerable , needs_answer , answered , or rejected .The possible question equivalence class state values are: incoming , or published . |
{"state": ["State_Value"]}
|
not_associated_with
|
question | A question equivalence class ID filter that allows you to find questions that are not in a particular question equivalence class. You can use this option to check whether a particular question equivalence rule (which you specify in the text filter) matches questions that are in another question equivalence class, or that are not associated with a question equivalence class. |
{ "not_associated_with": [
|
likely_answer_for
|
question_equivalence_class | A question ID filter that allows you to find question equivalence classes that are likely to provide an answer for a particular question. This option finds the question equivalence classes that Answer Server considers to be most relevant when it calculates the likelihood of answer score (see Sort). | { "likely_answer_for": [
|
ignore_special_characters
|
question
question_equivalence_class |
A Boolean value that specifies whether to treat special query syntax characters and operators as syntax or as plain text. Set this parameter to true to ignore special characters in the text , answer_text , and reference_question_text properties, and treat the string as plain text. In this case, Answer Server ignores syntax characters, such as wildcards (? and *), as well as Boolean and proximity operators. The default value is false. |
{ "ignore_special_characters": true }
|
fieldtext
|
question
question_equivalence_class rule_suggestion |
A filter that contains additional FieldText to add to the query to your Answer Bank Agentstore. You can use this option to filter the resources that you retrieve to a subset of your Answer Bank data, without having to set up multiple Answer Bank systems and Agentstores. | { "fieldtext": "MATCH{France}:REGION" }
|
typeahead
|
question_equivalence_class | A filter that contains user text to expand with suggestions from the question equivalence classes in your Answer Bank. IDOL Query Manipulation Server (QMS) uses this value for Answer Bank typeahead expansion. Set the typeahead property to the user text. GetResources returns any question equivalence classes that expand from the user text. |
{ "typeahead": "why is the sky"}
|
typeahead_expansion_mode
|
question_equivalence_class | The mode to use when using Answer Bank for typeahead expansions. This option is valid only when you include the typeahead option. IDOL Query Manipulation Server (QMS) uses this value for Answer Bank typeahead expansion. The default value is trailing, which provides suggestions that start with the text that you provide. Set it to substring to return any suggestion that contain the text that you provide as a substring. |
{ "typeahead_expansion_mode": "substring" }
|
You can specify multiple filters in the same JSON object.
For example, the following JSON object filters by the text President, and the question state incoming
, answerable
, or needs_answer
:
{
"text" : "President", "state" : [ "incoming", "answerable", "needs_answer" ] }
Actions: | GetResources
|
Type: | String |
Default: | all |
Example: | Filter=%7B%20%22text%22%3A%20%22some%20search%20text%22%20%7D
|
See Also: | IDs |