Google Cloud Natural Language

Overview

Google Cloud Natural Language tool for analyzing document sentiment. Supports both inline text content and Google Cloud Storage URIs as input sources. Returns overall document sentiment and per-sentence sentiment scores and magnitudes.

Node Type

Action — processes input items and produces output

Input / Output

Direction	Port(s)
Input	Input
Output	Output, Error

Credentials

This tool requires Google Cloud Natural Language OAuth2 API credentials. See the Credentials Guide for setup instructions.

Resources

Resource	Value
Document	document

Operations

Operation	Value	Description
Document: Analyze Sentiment	analyzeSentiment	Analyze sentiment of a document

Parameters

Document: Analyze Sentiment

Parameter	Type	Required	Default	Description
Source	options	Yes	content	The source of the document: a string containing the content or a Google Cloud Storage URI.
				Options: content, gcsContentUri (Google Cloud Storage URI)
Content	string	Yes	—	The content of the input in string format. Cloud audit logging exempt since it is based on user data. (shown when Source is content)
Google Cloud Storage URI	string	Yes	—	The Google Cloud Storage URI where the file content is located. This URI must be of the form: gs://bucket_name/object_name. (shown when Source is gcsContentUri)
Options	collection	No	{}	—
— Document Type	options	No	PLAIN_TEXT	The type of input document.
				Options: HTML, PLAIN_TEXT
— Encoding Type	options	No	UTF16	The encoding type used by the API to calculate sentence offsets.
				Options: NONE, UTF8, UTF16, UTF32
— Language	options	No	en	The language of the document (if not specified, the language is automatically detected). Both ISO and BCP-47 language codes are accepted.
				Options: ar (Arabic), zh (Chinese (Simplified)), zh-Hant (Chinese (Traditional)), nl (Dutch), en (English), fr (French), de (German), id (Indonesian), it (Italian), ja (Japanese), ko (Korean), pt (Portuguese (Brazilian & Continental)), es (Spanish), th (Thai), tr (Turkish), vi (Vietnamese)

All Operations

Parameter	Type	Required	Default	Description
Authentication	options	No	oAuth2	Authentication method to use.
				Options: oAuth2 (OAuth2 (recommended)), serviceAccount
Google Account	credential	No	—	Connect or select your Google account. (shown when Authentication is oAuth2)
Service Account Email	string	Yes	—	The email address of the Google service account. (shown when Authentication is serviceAccount)
Private Key	string	Yes	—	The private key from the service account JSON key file. (shown when Authentication is serviceAccount)
Max Concurrency	number	No	10	Maximum number of items to process concurrently.

Output Data

One output item per input item. The sentiment analysis response is merged onto the input item JSON, so the fields the item already carried stay addressable downstream. Binary data is not forwarded — attach files after this node, not before it.

The response is passed through exactly as the Google Cloud Natural Language API returns it: the overall document sentiment (a score and a magnitude), the per-sentence breakdown with a score and magnitude for each sentence, and the language the document was analysed in. Run the node once and inspect the output item to see the exact key names before you write expressions against it.

A sentiment score runs from -1 (negative) to 1 (positive). A magnitude is unbounded and grows with the amount of emotional content, so a long mixed document can have a score near zero and a high magnitude.

Example Configuration

Analyse the sentiment of text carried on the item:

{
  "type": "google_cloud_natural_language",
  "parameters": {
    "resource": "document",
    "operation": "analyzeSentiment",
    "source": "content",
    "content": "{{ $json.review }}"
  }
}

Analyse a document stored in Google Cloud Storage, in Spanish:

{
  "type": "google_cloud_natural_language",
  "parameters": {
    "resource": "document",
    "operation": "analyzeSentiment",
    "source": "gcsContentUri",
    "gcsContentUri": "gs://my-bucket/documents/review.txt",
    "options": {
      "documentType": "PLAIN_TEXT",
      "encodingType": "UTF16",
      "language": "es"
    }
  }
}

Analyse an HTML document and cap how many items run at once:

{
  "type": "google_cloud_natural_language",
  "parameters": {
    "resource": "document",
    "operation": "analyzeSentiment",
    "source": "content",
    "content": "{{ $json.html }}",
    "options": {
      "documentType": "HTML",
      "encodingType": "UTF8",
      "language": "en"
    },
    "maxConcurrency": 5
  }
}

Error Handling

Mode	Behavior
stop	Halts workflow on first error
continue	Skips failed items, passes successful ones through
errorPort	Routes failed items to Error output port

Tips

This tool analyzes the sentiment of text documents using the Google Cloud Natural Language service. It is ideal for determining the overall emotional tone of a document or the sentiment of individual sentences within it. The tool produces overall document sentiment, including scores and magnitudes, and per-sentence sentiment scores.

Behavior notes

• Pick the right document type. Choose HTML when the content is markup and the API will ignore the tags; leave it on PLAIN_TEXT and the tags are analysed as words.
• Encoding type only affects offsets. It decides how the character offsets in the per-sentence results are counted, not how the text is read. Leave it at UTF16 unless a downstream node slices the original string using those offsets.
• Leave Language unset to auto-detect. Set it explicitly only when you already know the language — a wrong value degrades the scores.
• One item per document. To score many documents, produce one item per document upstream (a Split Out node works well) rather than concatenating them into one string.