Azure Cognitive Service for Language

Azure Cognitive Service for Language, previously known as 'Text Analytics' connector detects language, sentiment and more of the text you provide.

This connector is available in the following products and regions:

Service Class Regions
Logic Apps Standard All Logic Apps regions
Power Automate Standard All Power Automate regions
Power Apps - Not available
Contact
Name Microsoft
URL https://gallery.cortanaanalytics.com/MachineLearningAPI/Text-Analytics-2
Email mlapi@microsoft.com
Connector Metadata
Publisher Microsoft
Website https://azure.microsoft.com/services/cognitive-services/text-analytics/

Note

This connector was previously known as "Text Analytics" or "Microsoft Cognitive Services Text Analytics".

To use this integration, you will need a Cognitive Service resource in the Azure portal. You will get an endpoint and a key for authenticating your applications. To make a connection, provide the Account key, site URL and select Create connection. For operation costs on your connection, learn more here.

You're now ready to start using this integration.

Connecting with Azure Active Directory

To create connections with AAD, you must add the Cognitive Services User role to your account to access. To use this connection, you must input the unique subdomain associated wih your Cognitive Services resource into the supporting operations.

Operations not supported with AAD authentication

When using AAD authentication only V4 actions are supported. Deprecated actions will continue to work with Access Key authentication, but will fail if used with an AAD connection.

Creating a connection

The connector supports the following authentication types:

Api Key ApiKey All regions Shareable
Default [DEPRECATED] This option is only for older connections without an explicit authentication type, and is only provided for backward compatibility. All regions Not shareable

Api Key

Auth ID: keyBasedAuth

Applicable: All regions

ApiKey

This is shareable connection. If the power app is shared with another user, connection is shared as well. For more information, please see the Connectors overview for canvas apps - Power Apps | Microsoft Docs

Name Type Description Required
Account Key securestring Cognitive Services Account Key True
Site URL string Root site url (Example: https://westus.api.cognitive.microsoft.com)

Default [DEPRECATED]

Applicable: All regions

This option is only for older connections without an explicit authentication type, and is only provided for backward compatibility.

This is not shareable connection. If the power app is shared with another user, another user will be prompted to create new connection explicitly.

Name Type Description Required
Account Key securestring Cognitive Services Account Key True
Site URL string Root site url (Example: https://westus.api.cognitive.microsoft.com). If not specified, site url will be defaulted to https://westus.api.cognitive.microsoft.com.

Throttling Limits

Name Calls Renewal Period
API calls per connection120060 seconds

Actions

Detect Entities (V2) [DEPRECATED]

This action has been deprecated. Please use Named Entity Recognition (V4) instead.

Returns a list of known entities and general named entities ("Person", "Location", "Organization" etc) in a given document.

Detect Language (V2) [DEPRECATED]

This action has been deprecated. Please use Detect Language (V4) instead.

Returns the detected language and a numeric score between 0 and 1 for each document. Scores close to 1 indicate 100% certainty that the identified language is true. A total of 120 languages are supported.

Detect Language (V3.0)

The API returns the detected language and a numeric score between 0 and 1. Scores close to 1 indicate 100% certainty that the identified language is true. See the Language support for Language Detection for the list of supported languages.

Detect Language (V3.1) (Preview)

The API returns the detected language and a numeric score between 0 and 1. Scores close to 1 indicate 100% certainty that the identified language is true. See the Language support for Language Detection for the list of supported languages.

Detect Language (V4) (Preview)

The API returns the detected language and a numeric score between 0 and 1. Scores close to 1 indicate 100% certainty that the identified language is true. See the Language support for Language Detection for the list of supported languages.

Detect Language [DEPRECATED]

This action has been deprecated. Please use Detect Language (V4) instead.

Returns the detected language and a numeric score between 0 and 1. Scores close to 1 indicate 100% certainty that the identified language is true. A total of 120 languages are supported.

Detect Language [DEPRECATED]

This action has been deprecated. Please use Detect Language (V4) instead.

Returns the detected language and a numeric score between 0 and 1. Scores close to 1 indicate 100% certainty that the identified language is true. A total of 120 languages are supported.

Detect Personal Information (V3.1)

The API returns a list of entities with personal information ("SSN", "Bank Account" etc) in the document. For the list of supported entity types, check Supported Personally Identifiable Information (PII) entity categories. See the Supported languages in Azure Cognitive Service for Language for the list of supported languages.

Detect Personal Information (V4) (Preview)

The API returns a list of entities with personal information ("SSN", "Bank Account" etc) in the document. For the list of supported entity types, check Supported Personally Identifiable Information (PII) entity categories. See the Supported languages in Azure Cognitive Service for Language for the list of supported languages.

Detect Sentiment (V2) [DEPRECATED]

This action has been deprecated. Please use Sentiment (V4) instead.

Returns a numeric score between 0 and 1 for each document. Scores close to 1 indicate positive sentiment, while scores close to 0 indicate negative sentiment.

Detect Sentiment [DEPRECATED]

This action has been deprecated. Please use Sentiment (V4) instead.

Returns a numeric score between 0 and 1. Scores close to 1 indicate positive sentiment, while scores close to 0 indicate negative sentiment.

Detect Sentiment [DEPRECATED]

This action has been deprecated. Please use Sentiment (V4) instead.

Returns a numeric score between 0 and 1. Scores close to 1 indicate positive sentiment, while scores close to 0 indicate negative sentiment. Currently, the following languages are supported: English, Spanish, French, Portuguese.

Entities [DEPRECATED]

This action has been deprecated. Please use Named Entity Recognition (V4) instead.

Returns a list of known entities and general named entities ("Person", "Location", "Organization" etc) in a given document.

Entity Linking (V3.0)

The API returns a list of recognized entities with links to a well-known knowledge base. See the language support for the list of supported languages.

Entity Linking (V3.1) (Preview)

The API returns a list of recognized entities with links to a well known knowledge base. See the language support for the list of supported languages.

Entity Linking (V4) (Preview)

The API returns a list of recognized entities with links to a well known knowledge base. See the language support for the list of supported languages.

Key Phrases (V2) [DEPRECATED]

This action has been deprecated. Please use Key Phrases (V4) instead.

Returns a list of strings denoting the key talking points in the input text for each document.

Key Phrases (V3.0)

The API returns a list of strings denoting the key phrases in the input text. See the Language support for Key Phrase Extraction for the list of supported languages.

Key Phrases (V3.1) (Preview)

The API returns a list of strings denoting the key phrases in the input text. See the Language support for Key Phrase Extraction for the list of supported languages.

Key Phrases (V4) (Preview)

The API returns a list of strings denoting the key phrases in the input text. See the Language support for Key Phrase Extraction for the list of supported languages.

Key Phrases [DEPRECATED]

This action has been deprecated. Please use Key Phrases (V4) instead.

Returns a list of strings denoting the key talking points in the input text.

Key Phrases [DEPRECATED]

This action has been deprecated. Please use Key Phrases (V4) instead.

Returns a list of strings denoting the key talking points in the input text. Currently, the following languages are supported: English, German, Spanish and Japanese.

Named Entity Recognition (V3.0)

The API returns a list of general named entities in a given document. For the list of supported entity types, check Supported Named Entity Recognition (NER) entity categories.

Named Entity Recognition (V3.1) (Preview)

The API returns a list of general named entities in a given document. For the list of supported entity types, check Supported Named Entity Recognition (NER) entity categories.

Named Entity Recognition (V4) (Preview)

The API returns a list of general named entities in a given document. For the list of supported entity types, check Supported Named Entity Recognition (NER) entity categories.

Sentiment (V3.0)

The API returns a sentiment prediction, as well as sentiment scores for each sentiment class (Positive, Negative, and Neutral) for the document and each sentence within it. See the Sentiment Analysis and Opinion Mining language support for the list of supported languages.

Sentiment (V3.1) (Preview)

The API returns a detailed sentiment analysis for the input text. The analysis is done in multiple levels of granularity, start from the a document level, down to sentence and key terms (targets and assessments).

Sentiment (V4) (Preview)

The API returns a detailed sentiment analysis for the input text. The analysis is done in multiple levels of granularity, start from the a document level, down to sentence and key terms (targets and assessments).

Detect Entities (V2) [DEPRECATED]

This action has been deprecated. Please use Named Entity Recognition (V4) instead.

Returns a list of known entities and general named entities ("Person", "Location", "Organization" etc) in a given document.

Parameters

Name Key Required Type Description
Id
id True string

Unique, non-empty document identifier.

Text
text True string

The text to analyze.

Language
language string

This is the 2 letter ISO 639-1 representation of a language. For example, use "en" for English; "es" for Spanish etc.,

Returns

Detect Language (V2) [DEPRECATED]

This action has been deprecated. Please use Detect Language (V4) instead.

Returns the detected language and a numeric score between 0 and 1 for each document. Scores close to 1 indicate 100% certainty that the identified language is true. A total of 120 languages are supported.

Parameters

Name Key Required Type Description
Id
id True string

Unique, non-empty document identifier.

Text
text True string

The text to analyze.

Returns

Detect Language (V3.0)

The API returns the detected language and a numeric score between 0 and 1. Scores close to 1 indicate 100% certainty that the identified language is true. See the Language support for Language Detection for the list of supported languages.

Parameters

Name Key Required Type Description
Model Version
model-version string

(Optional) This value indicates which model will be used for scoring. If a model-version is not specified, the API should default to the latest, non-preview version.

Show Statistics
showStats boolean

(Optional) if set to true, response will contain input and document level statistics.

id
id True string

Unique, non-empty document identifier.

text
text True string
countryHint
countryHint string

Returns

Detect Language (V3.1) (Preview)

The API returns the detected language and a numeric score between 0 and 1. Scores close to 1 indicate 100% certainty that the identified language is true. See the Language support for Language Detection for the list of supported languages.

Parameters

Name Key Required Type Description
model-version
model-version string

(Optional) This value indicates which model will be used for scoring. If a model-version is not specified, the API should default to the latest, non-preview version.

showStats
showStats boolean

(Optional) if set to true, response will contain request and document level statistics.

loggingOptOut
loggingOptOut boolean

(Optional) If set to true, you opt-out of having your text input logged for troubleshooting. By default, Azure Cognitive Service for Language logs your input text for 48 hours, solely to allow for troubleshooting issues in providing you with the Azure Cognitive Service for Language natural language processing functions. Setting this parameter to true, disables input logging and may limit our ability to remediate issues that occur. Please see Cognitive Services Compliance and Privacy notes at https://aka.ms/cs-compliance for additional details, and Microsoft Responsible AI principles at https://www.microsoft.com/en-us/ai/responsible-ai.

id
id True string

Unique, non-empty document identifier.

text
text True string
countryHint
countryHint string

Returns

Detect Language (V4) (Preview)

The API returns the detected language and a numeric score between 0 and 1. Scores close to 1 indicate 100% certainty that the identified language is true. See the Language support for Language Detection for the list of supported languages.

Parameters

Name Key Required Type Description
Resource Subdomain or Region
subdomain True string

Azure subdomain or geographic region (e.g. westus)

model-version
model-version string

(Optional) This value indicates which model will be used for scoring. If a model-version is not specified, the API should default to the latest, non-preview version.

showStats
showStats boolean

(Optional) if set to true, response will contain request and document level statistics.

loggingOptOut
loggingOptOut boolean

(Optional) If set to true, you opt-out of having your text input logged for troubleshooting. By default, Azure Cognitive Service for Language logs your input text for 48 hours, solely to allow for troubleshooting issues in providing you with the Azure Cognitive Service for Language natural language processing functions. Setting this parameter to true, disables input logging and may limit our ability to remediate issues that occur. Please see Cognitive Services Compliance and Privacy notes at https://aka.ms/cs-compliance for additional details, and Microsoft Responsible AI principles at https://www.microsoft.com/en-us/ai/responsible-ai.

id
id True string

Unique, non-empty document identifier.

text
text True string
countryHint
countryHint string

Returns

Detect Language [DEPRECATED]

This action has been deprecated. Please use Detect Language (V4) instead.

Returns the detected language and a numeric score between 0 and 1. Scores close to 1 indicate 100% certainty that the identified language is true. A total of 120 languages are supported.

Parameters

Name Key Required Type Description
Text
text string

The text to analyze.

Number of languages
numberOfLanguagesToDetect integer

The number of languages to detect. Set to 1 by default.

Returns

Detect Language [DEPRECATED]

This action has been deprecated. Please use Detect Language (V4) instead.

Returns the detected language and a numeric score between 0 and 1. Scores close to 1 indicate 100% certainty that the identified language is true. A total of 120 languages are supported.

Parameters

Name Key Required Type Description
Text
text string

The text to analyze.

Returns

Detect Personal Information (V3.1)

The API returns a list of entities with personal information ("SSN", "Bank Account" etc) in the document. For the list of supported entity types, check Supported Personally Identifiable Information (PII) entity categories. See the Supported languages in Azure Cognitive Service for Language for the list of supported languages.

Parameters

Name Key Required Type Description
model-version
model-version string

(Optional) This value indicates which model will be used for scoring. If a model-version is not specified, the API should default to the latest, non-preview version.

showStats
showStats boolean

(Optional) if set to true, response will contain request and document level statistics.

loggingOptOut
loggingOptOut boolean

(Optional) If set to true, you opt-out of having your text input logged for troubleshooting. By default, Azure Cognitive Service for Language logs your input text for 48 hours, solely to allow for troubleshooting issues in providing you with the Azure Cognitive Service for Language natural language processing functions. Setting this parameter to true, disables input logging and may limit our ability to remediate issues that occur. Please see Cognitive Services Compliance and Privacy notes at https://aka.ms/cs-compliance for additional details, and Microsoft Responsible AI principles at https://www.microsoft.com/en-us/ai/responsible-ai.

domain
domain string

(Optional) if specified, will set the PII domain to include only a subset of the entity categories. Possible values include: 'PHI', 'none'.

stringIndexType
stringIndexType string

(Optional) Specifies the method used to interpret string offsets. Defaults to Text Elements (Graphemes) according to Unicode v8.0.0. For additional information see https://aka.ms/text-analytics-offsets

piiCategories
piiCategories array

(Optional) describes the PII categories to return

id
id True string

A unique, non-empty document identifier.

text
text True string

The input text to process.

language
language string

(Optional) This is the 2 letter ISO 639-1 representation of a language. For example, use "en" for English; "es" for Spanish etc. If not set, use "en" for English as default.

Returns

Body
PiiResult

Detect Personal Information (V4) (Preview)

The API returns a list of entities with personal information ("SSN", "Bank Account" etc) in the document. For the list of supported entity types, check Supported Personally Identifiable Information (PII) entity categories. See the Supported languages in Azure Cognitive Service for Language for the list of supported languages.

Parameters

Name Key Required Type Description
Resource Subdomain or Region
subdomain True string

Azure subdomain or geographic region (e.g. westus)

model-version
model-version string

(Optional) This value indicates which model will be used for scoring. If a model-version is not specified, the API should default to the latest, non-preview version.

showStats
showStats boolean

(Optional) if set to true, response will contain request and document level statistics.

loggingOptOut
loggingOptOut boolean

(Optional) If set to true, you opt-out of having your text input logged for troubleshooting. By default, Azure Cognitive Service for Language logs your input text for 48 hours, solely to allow for troubleshooting issues in providing you with the Azure Cognitive Service for Language natural language processing functions. Setting this parameter to true, disables input logging and may limit our ability to remediate issues that occur. Please see Cognitive Services Compliance and Privacy notes at https://aka.ms/cs-compliance for additional details, and Microsoft Responsible AI principles at https://www.microsoft.com/en-us/ai/responsible-ai.

domain
domain string

(Optional) if specified, will set the PII domain to include only a subset of the entity categories. Possible values include: 'PHI', 'none'.

stringIndexType
stringIndexType string

(Optional) Specifies the method used to interpret string offsets. Defaults to Text Elements (Graphemes) according to Unicode v8.0.0. For additional information see https://aka.ms/text-analytics-offsets

piiCategories
piiCategories array

(Optional) describes the PII categories to return

id
id True string

A unique, non-empty document identifier.

text
text True string

The input text to process.

language
language string

(Optional) This is the 2 letter ISO 639-1 representation of a language. For example, use "en" for English; "es" for Spanish etc. If not set, use "en" for English as default.

Returns

Body
PiiResult

Detect Sentiment (V2) [DEPRECATED]

This action has been deprecated. Please use Sentiment (V4) instead.

Returns a numeric score between 0 and 1 for each document. Scores close to 1 indicate positive sentiment, while scores close to 0 indicate negative sentiment.

Parameters

Name Key Required Type Description
Id
id True string

Unique, non-empty document identifier.

Text
text True string

The text to analyze.

Language
language string

This is the 2 letter ISO 639-1 representation of a language. For example, use "en" for English; "es" for Spanish etc.,

Returns

Detect Sentiment [DEPRECATED]

This action has been deprecated. Please use Sentiment (V4) instead.

Returns a numeric score between 0 and 1. Scores close to 1 indicate positive sentiment, while scores close to 0 indicate negative sentiment.

Parameters

Name Key Required Type Description
Text
text string

The text to analyze.

Language
language string

This is the 2 letter ISO 639-1 representation of a language. For example, use "en" for English; "es" for Spanish etc.,

Returns

Detect Sentiment [DEPRECATED]

This action has been deprecated. Please use Sentiment (V4) instead.

Returns a numeric score between 0 and 1. Scores close to 1 indicate positive sentiment, while scores close to 0 indicate negative sentiment. Currently, the following languages are supported: English, Spanish, French, Portuguese.

Parameters

Name Key Required Type Description
Text
text string

The text to analyze.

Language
language string

This is the 2 letter ISO 639-1 representation of a language. For example, use "en" for English; "es" for Spanish etc.,

Returns

Entities [DEPRECATED]

This action has been deprecated. Please use Named Entity Recognition (V4) instead.

Returns a list of known entities and general named entities ("Person", "Location", "Organization" etc) in a given document.

Parameters

Name Key Required Type Description
Text
text string

The text to analyze.

Language
language string

This is the 2 letter ISO 639-1 representation of a language. For example, use "en" for English; "es" for Spanish etc.,

Returns

Entity Linking (V3.0)

The API returns a list of recognized entities with links to a well-known knowledge base. See the language support for the list of supported languages.

Parameters

Name Key Required Type Description
Model Version
model-version string

(Optional) This value indicates which model will be used for scoring. If a model-version is not specified, the API should default to the latest, non-preview version.

Show Statistics
showStats boolean

(Optional) if set to true, response will contain input and document level statistics.

id
id True string

A unique, non-empty document identifier.

text
text True string

The input text to process.

language
language string

(Optional) This is the 2 letter ISO 639-1 representation of a language. For example, use "en" for English; "es" for Spanish etc. If not set, use "en" for English as default.

Returns

Entity Linking (V3.1) (Preview)

The API returns a list of recognized entities with links to a well known knowledge base. See the language support for the list of supported languages.

Parameters

Name Key Required Type Description
model-version
model-version string

(Optional) This value indicates which model will be used for scoring. If a model-version is not specified, the API should default to the latest, non-preview version.

showStats
showStats boolean

(Optional) if set to true, response will contain request and document level statistics.

loggingOptOut
loggingOptOut boolean

(Optional) If set to true, you opt-out of having your text input logged for troubleshooting. By default, Azure Cognitive Service for Language logs your input text for 48 hours, solely to allow for troubleshooting issues in providing you with the Azure Cognitive Service for Language natural language processing functions. Setting this parameter to true, disables input logging and may limit our ability to remediate issues that occur. Please see Cognitive Services Compliance and Privacy notes at https://aka.ms/cs-compliance for additional details, and Microsoft Responsible AI principles at https://www.microsoft.com/en-us/ai/responsible-ai.

stringIndexType
stringIndexType string

(Optional) Specifies the method used to interpret string offsets. Defaults to Text Elements (Graphemes) according to Unicode v8.0.0. For additional information see https://aka.ms/text-analytics-offsets

id
id True string

A unique, non-empty document identifier.

text
text True string

The input text to process.

language
language string

(Optional) This is the 2 letter ISO 639-1 representation of a language. For example, use "en" for English; "es" for Spanish etc. If not set, use "en" for English as default.

Returns

Entity Linking (V4) (Preview)

The API returns a list of recognized entities with links to a well known knowledge base. See the language support for the list of supported languages.

Parameters

Name Key Required Type Description
Resource Subdomain or Region
subdomain True string

Azure subdomain or geographic region (e.g. westus)

model-version
model-version string

(Optional) This value indicates which model will be used for scoring. If a model-version is not specified, the API should default to the latest, non-preview version.

showStats
showStats boolean

(Optional) if set to true, response will contain request and document level statistics.

loggingOptOut
loggingOptOut boolean

(Optional) If set to true, you opt-out of having your text input logged for troubleshooting. By default, Azure Cognitive Service for Language logs your input text for 48 hours, solely to allow for troubleshooting issues in providing you with the Azure Cognitive Service for Language natural language processing functions. Setting this parameter to true, disables input logging and may limit our ability to remediate issues that occur. Please see Cognitive Services Compliance and Privacy notes at https://aka.ms/cs-compliance for additional details, and Microsoft Responsible AI principles at https://www.microsoft.com/en-us/ai/responsible-ai.

stringIndexType
stringIndexType string

(Optional) Specifies the method used to interpret string offsets. Defaults to Text Elements (Graphemes) according to Unicode v8.0.0. For additional information see https://aka.ms/text-analytics-offsets

id
id True string

A unique, non-empty document identifier.

text
text True string

The input text to process.

language
language string

(Optional) This is the 2 letter ISO 639-1 representation of a language. For example, use "en" for English; "es" for Spanish etc. If not set, use "en" for English as default.

Returns

Key Phrases (V2) [DEPRECATED]

This action has been deprecated. Please use Key Phrases (V4) instead.

Returns a list of strings denoting the key talking points in the input text for each document.

Parameters

Name Key Required Type Description
Id
id True string

Unique, non-empty document identifier.

Text
text True string

The text to analyze.

Language
language string

This is the 2 letter ISO 639-1 representation of a language. For example, use "en" for English; "es" for Spanish etc.,

Returns

Key Phrases (V3.0)

The API returns a list of strings denoting the key phrases in the input text. See the Language support for Key Phrase Extraction for the list of supported languages.

Parameters

Name Key Required Type Description
Model Version
model-version string

(Optional) This value indicates which model will be used for scoring. If a model-version is not specified, the API should default to the latest, non-preview version.

Show Statistics
showStats boolean

(Optional) if set to true, response will contain input and document level statistics.

id
id True string

A unique, non-empty document identifier.

text
text True string

The input text to process.

language
language string

(Optional) This is the 2 letter ISO 639-1 representation of a language. For example, use "en" for English; "es" for Spanish etc. If not set, use "en" for English as default.

Returns

Key Phrases (V3.1) (Preview)

The API returns a list of strings denoting the key phrases in the input text. See the Language support for Key Phrase Extraction for the list of supported languages.

Parameters

Name Key Required Type Description
model-version
model-version string

(Optional) This value indicates which model will be used for scoring. If a model-version is not specified, the API should default to the latest, non-preview version.

showStats
showStats boolean

(Optional) if set to true, response will contain request and document level statistics.

loggingOptOut
loggingOptOut boolean

(Optional) If set to true, you opt-out of having your text input logged for troubleshooting. By default, Azure Cognitive Service for Language logs your input text for 48 hours, solely to allow for troubleshooting issues in providing you with the Azure Cognitive Service for Language natural language processing functions. Setting this parameter to true, disables input logging and may limit our ability to remediate issues that occur. Please see Cognitive Services Compliance and Privacy notes at https://aka.ms/cs-compliance for additional details, and Microsoft Responsible AI principles at https://www.microsoft.com/en-us/ai/responsible-ai.

id
id True string

A unique, non-empty document identifier.

text
text True string

The input text to process.

language
language string

(Optional) This is the 2 letter ISO 639-1 representation of a language. For example, use "en" for English; "es" for Spanish etc. If not set, use "en" for English as default.

Returns

Key Phrases (V4) (Preview)

The API returns a list of strings denoting the key phrases in the input text. See the Language support for Key Phrase Extraction for the list of supported languages.

Parameters

Name Key Required Type Description
Resource Subdomain or Region
subdomain True string

Azure subdomain or geographic region (e.g. westus)

model-version
model-version string

(Optional) This value indicates which model will be used for scoring. If a model-version is not specified, the API should default to the latest, non-preview version.

showStats
showStats boolean

(Optional) if set to true, response will contain request and document level statistics.

loggingOptOut
loggingOptOut boolean

(Optional) If set to true, you opt-out of having your text input logged for troubleshooting. By default, Azure Cognitive Service for Language logs your input text for 48 hours, solely to allow for troubleshooting issues in providing you with the Azure Cognitive Service for Language natural language processing functions. Setting this parameter to true, disables input logging and may limit our ability to remediate issues that occur. Please see Cognitive Services Compliance and Privacy notes at https://aka.ms/cs-compliance for additional details, and Microsoft Responsible AI principles at https://www.microsoft.com/en-us/ai/responsible-ai.

id
id True string

A unique, non-empty document identifier.

text
text True string

The input text to process.

language
language string

(Optional) This is the 2 letter ISO 639-1 representation of a language. For example, use "en" for English; "es" for Spanish etc. If not set, use "en" for English as default.

Returns

Key Phrases [DEPRECATED]

This action has been deprecated. Please use Key Phrases (V4) instead.

Returns a list of strings denoting the key talking points in the input text.

Parameters

Name Key Required Type Description
Text
text string

The text to analyze.

Language
language string

This is the 2 letter ISO 639-1 representation of a language. For example, use "en" for English; "es" for Spanish etc.,

Returns

Key Phrases [DEPRECATED]

This action has been deprecated. Please use Key Phrases (V4) instead.

Returns a list of strings denoting the key talking points in the input text. Currently, the following languages are supported: English, German, Spanish and Japanese.

Parameters

Name Key Required Type Description
Text
text string

The text to analyze.

Language
language string

This is the 2 letter ISO 639-1 representation of a language. For example, use "en" for English; "es" for Spanish etc.,

Returns

Named Entity Recognition (V3.0)

The API returns a list of general named entities in a given document. For the list of supported entity types, check Supported Named Entity Recognition (NER) entity categories.

Parameters

Name Key Required Type Description
Model Version
model-version string

(Optional) This value indicates which model will be used for scoring. If a model-version is not specified, the API should default to the latest, non-preview version.

Show Statistics
showStats boolean

(Optional) if set to true, response will contain input and document level statistics.

id
id True string

A unique, non-empty document identifier.

text
text True string

The input text to process.

language
language string

(Optional) This is the 2 letter ISO 639-1 representation of a language. For example, use "en" for English; "es" for Spanish etc. If not set, use "en" for English as default.

Returns

Named Entity Recognition (V3.1) (Preview)

The API returns a list of general named entities in a given document. For the list of supported entity types, check Supported Named Entity Recognition (NER) entity categories.

Parameters

Name Key Required Type Description
model-version
model-version string

(Optional) This value indicates which model will be used for scoring. If a model-version is not specified, the API should default to the latest, non-preview version.

showStats
showStats boolean

(Optional) if set to true, response will contain request and document level statistics.

loggingOptOut
loggingOptOut boolean

(Optional) If set to true, you opt-out of having your text input logged for troubleshooting. By default, Azure Cognitive Service for Language logs your input text for 48 hours, solely to allow for troubleshooting issues in providing you with the Azure Cognitive Service for Language natural language processing functions. Setting this parameter to true, disables input logging and may limit our ability to remediate issues that occur. Please see Cognitive Services Compliance and Privacy notes at https://aka.ms/cs-compliance for additional details, and Microsoft Responsible AI principles at https://www.microsoft.com/en-us/ai/responsible-ai.

stringIndexType
stringIndexType string

(Optional) Specifies the method used to interpret string offsets. Defaults to Text Elements (Graphemes) according to Unicode v8.0.0. For additional information see https://aka.ms/text-analytics-offsets

id
id True string

A unique, non-empty document identifier.

text
text True string

The input text to process.

language
language string

(Optional) This is the 2 letter ISO 639-1 representation of a language. For example, use "en" for English; "es" for Spanish etc. If not set, use "en" for English as default.

Returns

Named Entity Recognition (V4) (Preview)

The API returns a list of general named entities in a given document. For the list of supported entity types, check Supported Named Entity Recognition (NER) entity categories.

Parameters

Name Key Required Type Description
Resource Subdomain or Region
subdomain True string

Azure subdomain or geographic region (e.g. westus)

model-version
model-version string

(Optional) This value indicates which model will be used for scoring. If a model-version is not specified, the API should default to the latest, non-preview version.

showStats
showStats boolean

(Optional) if set to true, response will contain request and document level statistics.

loggingOptOut
loggingOptOut boolean

(Optional) If set to true, you opt-out of having your text input logged for troubleshooting. By default, Azure Cognitive Service for Language logs your input text for 48 hours, solely to allow for troubleshooting issues in providing you with the Azure Cognitive Service for Language natural language processing functions. Setting this parameter to true, disables input logging and may limit our ability to remediate issues that occur. Please see Cognitive Services Compliance and Privacy notes at https://aka.ms/cs-compliance for additional details, and Microsoft Responsible AI principles at https://www.microsoft.com/en-us/ai/responsible-ai.

stringIndexType
stringIndexType string

(Optional) Specifies the method used to interpret string offsets. Defaults to Text Elements (Graphemes) according to Unicode v8.0.0. For additional information see https://aka.ms/text-analytics-offsets

id
id True string

A unique, non-empty document identifier.

text
text True string

The input text to process.

language
language string

(Optional) This is the 2 letter ISO 639-1 representation of a language. For example, use "en" for English; "es" for Spanish etc. If not set, use "en" for English as default.

Returns

Sentiment (V3.0)

The API returns a sentiment prediction, as well as sentiment scores for each sentiment class (Positive, Negative, and Neutral) for the document and each sentence within it. See the Sentiment Analysis and Opinion Mining language support for the list of supported languages.

Parameters

Name Key Required Type Description
Model Version
model-version string

(Optional) This value indicates which model will be used for scoring. If a model-version is not specified, the API should default to the latest, non-preview version.

Show Statistics
showStats boolean

(Optional) if set to true, response will contain input and document level statistics.

id
id True string

A unique, non-empty document identifier.

text
text True string

The input text to process.

language
language string

(Optional) This is the 2 letter ISO 639-1 representation of a language. For example, use "en" for English; "es" for Spanish etc. If not set, use "en" for English as default.

Returns

Sentiment (V3.1) (Preview)

The API returns a detailed sentiment analysis for the input text. The analysis is done in multiple levels of granularity, start from the a document level, down to sentence and key terms (targets and assessments).

Parameters

Name Key Required Type Description
model-version
model-version string

(Optional) This value indicates which model will be used for scoring. If a model-version is not specified, the API should default to the latest, non-preview version.

showStats
showStats boolean

(Optional) if set to true, response will contain request and document level statistics.

loggingOptOut
loggingOptOut boolean

(Optional) If set to true, you opt-out of having your text input logged for troubleshooting. By default, Azure Cognitive Service for Language logs your input text for 48 hours, solely to allow for troubleshooting issues in providing you with the Azure Cognitive Service for Language natural language processing functions. Setting this parameter to true, disables input logging and may limit our ability to remediate issues that occur. Please see Cognitive Services Compliance and Privacy notes at https://aka.ms/cs-compliance for additional details, and Microsoft Responsible AI principles at https://www.microsoft.com/en-us/ai/responsible-ai.

opinionMining
opinionMining boolean

(Optional) if set to true, response will contain not only sentiment prediction but also opinion mining (aspect-based sentiment analysis) results.

stringIndexType
stringIndexType string

(Optional) Specifies the method used to interpret string offsets. Defaults to Text Elements (Graphemes) according to Unicode v8.0.0. For additional information see https://aka.ms/text-analytics-offsets

id
id True string

A unique, non-empty document identifier.

text
text True string

The input text to process.

language
language string

(Optional) This is the 2 letter ISO 639-1 representation of a language. For example, use "en" for English; "es" for Spanish etc. If not set, use "en" for English as default.

Returns

Sentiment (V4) (Preview)

The API returns a detailed sentiment analysis for the input text. The analysis is done in multiple levels of granularity, start from the a document level, down to sentence and key terms (targets and assessments).

Parameters

Name Key Required Type Description
Resource Subdomain or Region
subdomain True string

Azure subdomain or geographic region (e.g. westus)

model-version
model-version string

(Optional) This value indicates which model will be used for scoring. If a model-version is not specified, the API should default to the latest, non-preview version.

showStats
showStats boolean

(Optional) if set to true, response will contain request and document level statistics.

loggingOptOut
loggingOptOut boolean

(Optional) If set to true, you opt-out of having your text input logged for troubleshooting. By default, Azure Cognitive Service for Language logs your input text for 48 hours, solely to allow for troubleshooting issues in providing you with the Azure Cognitive Service for Language natural language processing functions. Setting this parameter to true, disables input logging and may limit our ability to remediate issues that occur. Please see Cognitive Services Compliance and Privacy notes at https://aka.ms/cs-compliance for additional details, and Microsoft Responsible AI principles at https://www.microsoft.com/en-us/ai/responsible-ai.

opinionMining
opinionMining boolean

(Optional) if set to true, response will contain not only sentiment prediction but also opinion mining (aspect-based sentiment analysis) results.

stringIndexType
stringIndexType string

(Optional) Specifies the method used to interpret string offsets. Defaults to Text Elements (Graphemes) according to Unicode v8.0.0. For additional information see https://aka.ms/text-analytics-offsets

id
id True string

A unique, non-empty document identifier.

text
text True string

The input text to process.

language
language string

(Optional) This is the 2 letter ISO 639-1 representation of a language. For example, use "en" for English; "es" for Spanish etc. If not set, use "en" for English as default.

Returns

Definitions

DocumentError

Name Path Type Description
id
id string

Document Id.

error
error TextAnalyticsError

TextAnalyticsError

Name Path Type Description
code
code string

Error code.

message
message string

Error message.

target
target string

Error target.

innererror
innererror InnerError
details
details array of TextAnalyticsError

Details about specific errors that led to this reported error.

TextAnalyticsWarning

Name Path Type Description
code
code string

Error code.

message
message string

Warning message.

targetRef
targetRef string

A JSON pointer reference indicating the target object.

InnerError

Name Path Type Description
code
code string

Error code.

message
message string

Error message.

details
details object

Error details.

target
target string

Error target.

innererror
innererror InnerError

SentimentResponse

Name Path Type Description
documents
documents array of DocumentSentiment

Sentiment analysis per document.

errors
errors array of DocumentError

Errors by document id.

statistics
statistics RequestStatistics

if showStats=true was specified in the request this field will contain information about the request payload.

modelVersion
modelVersion string

This field indicates which model is used for scoring.

DocumentSentiment

Name Path Type Description
id
id string

Unique, non-empty document identifier.

sentiment
sentiment string

Predicted sentiment for document (Negative, Neutral, Positive, or Mixed).

statistics
statistics DocumentStatistics

if showStats=true was specified in the request this field will contain information about the document payload.

confidenceScores
confidenceScores SentimentConfidenceScorePerLabel

Represents the confidence scores between 0 and 1 across all sentiment classes: positive, neutral, negative.

sentences
sentences array of SentenceSentiment

Sentence level sentiment analysis.

warnings
warnings array of TextAnalyticsWarning

Warnings encountered while processing document.

RequestStatistics

if showStats=true was specified in the request this field will contain information about the request payload.

Name Path Type Description
documentsCount
documentsCount integer

Number of documents submitted in the request.

validDocumentsCount
validDocumentsCount integer

Number of valid documents. This excludes empty, over-size limit or non-supported languages documents.

erroneousDocumentsCount
erroneousDocumentsCount integer

Number of invalid documents. This includes empty, over-size limit or non-supported languages documents.

transactionsCount
transactionsCount integer

Number of transactions for the request.

DocumentStatistics

if showStats=true was specified in the request this field will contain information about the document payload.

Name Path Type Description
charactersCount
charactersCount integer

Number of text elements recognized in the document.

transactionsCount
transactionsCount integer

Number of transactions for the document.

SentimentConfidenceScorePerLabel

Represents the confidence scores between 0 and 1 across all sentiment classes: positive, neutral, negative.

Name Path Type Description
positive
positive double
neutral
neutral double
negative
negative double

SentenceSentiment

Name Path Type Description
text
text string

The sentence text.

sentiment
sentiment string

The predicted Sentiment for the sentence.

confidenceScores
confidenceScores SentimentConfidenceScorePerLabel

Represents the confidence scores between 0 and 1 across all sentiment classes: positive, neutral, negative.

offset
offset integer

The sentence offset from the start of the document.

length
length integer

The length of the sentence by Unicode standard.

DocumentEntities

Name Path Type Description
id
id string

Unique, non-empty document identifier.

entities
entities array of Entity

Recognized entities in the document.

warnings
warnings array of TextAnalyticsWarning

Warnings encountered while processing document.

statistics
statistics DocumentStatistics

if showStats=true was specified in the request this field will contain information about the document payload.

PiiDocumentEntities

Name Path Type Description
id
id string

Unique, non-empty document identifier.

redactedText
redactedText string

Returns redacted text.

entities
entities array of Entity

Recognized entities in the document.

warnings
warnings array of TextAnalyticsWarning

Warnings encountered while processing document.

statistics
statistics DocumentStatistics

if showStats=true was specified in the request this field will contain information about the document payload.

Entity

Name Path Type Description
text
text string

Entity text as appears in the request.

category
category string

Entity type, such as Person/Location/Org/SSN etc

subcategory
subcategory string

Entity sub type, such as Age/Year/TimeRange etc

offset
offset integer

Start position (in Unicode characters) for the entity text.

length
length integer

Length (in Unicode characters) for the entity text.

confidenceScore
confidenceScore double

Confidence score between 0 and 1 of the extracted entity.

EntityLinkingResult

Name Path Type Description
documents
documents array of DocumentLinkedEntities

Response by document

errors
errors array of DocumentError

Errors by document id.

statistics
statistics RequestStatistics

if showStats=true was specified in the request this field will contain information about the request payload.

modelVersion
modelVersion string

This field indicates which model is used for scoring.

DocumentLinkedEntities

Name Path Type Description
id
id string

Unique, non-empty document identifier.

entities
entities array of LinkedEntity

Recognized well-known entities in the document.

warnings
warnings array of TextAnalyticsWarning

Warnings encountered while processing document.

statistics
statistics DocumentStatistics

if showStats=true was specified in the request this field will contain information about the document payload.

LinkedEntity

Name Path Type Description
name
name string

Entity Linking formal name.

matches
matches array of Match

List of instances this entity appears in the text.

language
language string

Language used in the data source.

id
id string

Unique identifier of the recognized entity from the data source.

url
url string

URL for the entity's page from the data source.

dataSource
dataSource string

Data source used to extract entity linking, such as Wiki/Bing etc.

Match

Name Path Type Description
confidenceScore
confidenceScore double

If a well-known item is recognized, a decimal number denoting the confidence level between 0 and 1 will be returned.

text
text string

Entity text as appears in the request.

offset
offset integer

Start position (in Unicode characters) for the entity match text.

length
length integer

Length (in Unicode characters) for the entity match text.

DocumentKeyPhrases

Name Path Type Description
id
id string

Unique, non-empty document identifier.

keyPhrases
keyPhrases array of string

A list of representative words or phrases. The number of key phrases returned is proportional to the number of words in the input document.

warnings
warnings array of TextAnalyticsWarning

Warnings encountered while processing document.

statistics
statistics DocumentStatistics

if showStats=true was specified in the request this field will contain information about the document payload.

DocumentLanguage

Name Path Type Description
id
id string

Unique, non-empty document identifier.

detectedLanguage
detectedLanguage DetectedLanguage
warnings
warnings array of TextAnalyticsWarning

Warnings encountered while processing document.

statistics
statistics DocumentStatistics

if showStats=true was specified in the request this field will contain information about the document payload.

DetectedLanguage

Name Path Type Description
name
name string

Long name of a detected language (e.g. English, French).

iso6391Name
iso6391Name string

A two letter representation of the detected language according to the ISO 639-1 standard (e.g. en, fr).

confidenceScore
confidenceScore double

A confidence score between 0 and 1. Scores close to 1 indicate 100% certainty that the identified language is true.

KeyPhraseResultV3

Name Path Type Description
documents
documents array of DocumentKeyPhrases

Response by document

errors
errors array of DocumentError

Errors by document id.

statistics
statistics RequestStatistics

if showStats=true was specified in the request this field will contain information about the request payload.

modelVersion
modelVersion string

This field indicates which model is used for scoring.

EntitiesResultV3

Name Path Type Description
documents
documents array of DocumentEntities

Response by document

errors
errors array of DocumentError

Errors by document id.

statistics
statistics RequestStatistics

if showStats=true was specified in the request this field will contain information about the request payload.

modelVersion
modelVersion string

This field indicates which model is used for scoring.

PiiResult

Name Path Type Description
documents
documents array of PiiDocumentEntities

Response by document

errors
errors array of DocumentError

Errors by document id.

statistics
statistics RequestStatistics

if showStats=true was specified in the request this field will contain information about the request payload.

modelVersion
modelVersion string

This field indicates which model is used for scoring.

LanguageResultV3

Name Path Type Description
documents
documents array of DocumentLanguage

Response by document

errors
errors array of DocumentError

Errors by document id.

statistics
statistics RequestStatistics

if showStats=true was specified in the request this field will contain information about the request payload.

modelVersion
modelVersion string

This field indicates which model is used for scoring.

KeyPhraseResults

Name Path Type Description
documents
documents array of KeyPhraseResult

EntitiesResults

Name Path Type Description
documents
documents array of EntitiesResult

LanguageResults

Name Path Type Description
documents
documents array of LanguageResult

SentimentResults

Name Path Type Description
documents
documents array of SentimentResult

KeyPhraseResult

Name Path Type Description
Key Phrases
keyPhrases array of string

A list of representative words or phrases. The number of key phrases returned is proportional to the number of words in the input document.

Id
id string

The unique document identifier.

EntitiesResult

Name Path Type Description
Entities
entities array of object

A list of disambiguated entities with links to more information on the web (Wikipedia and Bing).

Name
entities.name string

Name of the entity.

Matches
entities.matches array of object

A list of matches for the entity.

items
entities.matches object
Wikipedia Language
entities.wikipediaLanguage string

Wikipedia language of the entity.

Wikipedia Id
entities.wikipediaId string

Wikipedia Id of the entity.

Wikipedia Url
entities.wikipediaUrl string

Wikipedia url of the entity.

Bing Id
entities.bingId string

Bing Id of the entity.

Type
entities.type string

Type of the entity.

Subtype
entities.subType string

Subtype of the entity.

Id
id string

The unique document identifier.

LanguageResult

Name Path Type Description
Id
id string

The unique document identifier.

detectedLanguages
detectedLanguages array of DetectedLanguageV2

A list of extracted languages.

SentimentResult

Name Path Type Description
Score
score double

A decimal number between 0 and 1 denoting the sentiment of the document. A score above 0.7 usually refers to a positive document while a score below 0.3 normally has a negative connotation. Mid values refer to neutral text.

Id
id string

The unique document identifier.

DetectedLanguageV2

Name Path Type Description
Name
name string

The long name of a detected language (e.g. English, French).

Language Code
iso6391Name string

A two letter representation of the detected language according to the ISO 639-1 standard (e.g. en, fr).

Score
score double

A confidence score between 0 and 1. Scores close to 1 indicate 100% certainty that the identified language is true.