Text Classifier
A built-in n8n node for large language model-based text classification. It performs zero-shot or few-shot classification through user-defined categories and descriptions, without the need to train a dedicated model.
Timeline (last 90 days)
In n8n workflows, the classification stability of the Text Classifier node depends on the specificity and boundary clarity of category names and descriptions; semantic overlap causes the model's judgments to waver.
Refunds and account security are high-risk scenarios where an inappropriate AI reply could trigger disputes, so this workflow routes such requests directly past the AI to humans.
The classifier's category division depends on prompt quality, and in actual use it needs repeated adjustment according to the business.
n8n's Text Classifier node sends the text to be classified together with predefined category descriptions to the underlying large language model and returns a structured category label that can be used directly for IF branch decisions.
This n8n workflow uses a text classifier to divide customer inquiries into two categories, "can draft" (can be drafted automatically) and "needs human" (requires human handling); login help and simple inquiries fall into the former, while unsubscribe, refund, and security-related issues fall into the latter.
The working mechanism of Text Classifier is essentially zero-shot/few-shot classification—the model does not need special training and only needs to read the category descriptions to make a judgment, so the clarity of the category descriptions directly determines classification accuracy.
n8n has hundreds of built-in integration nodes (Gmail, Slack, databases, HTTP requests, etc.) and also provides AI-related nodes such as Text Classifier and Agent.
All Facts (7)
In n8n workflows, the classification stability of the Text Classifier node depends on the specificity and boundary clarity of category names and descriptions; semantic overlap causes the model's judgments to waver.
50%UnverifiedRefunds and account security are high-risk scenarios where an inappropriate AI reply could trigger disputes, so this workflow routes such requests directly past the AI to humans.
50%UnverifiedThe classifier's category division depends on prompt quality, and in actual use it needs repeated adjustment according to the business.
50%Unverifiedn8n's Text Classifier node sends the text to be classified together with predefined category descriptions to the underlying large language model and returns a structured category label that can be used directly for IF branch decisions.
50%UnverifiedThis n8n workflow uses a text classifier to divide customer inquiries into two categories, "can draft" (can be drafted automatically) and "needs human" (requires human handling); login help and simple inquiries fall into the former, while unsubscribe, refund, and security-related issues fall into the latter.
50%UnverifiedThe working mechanism of Text Classifier is essentially zero-shot/few-shot classification—the model does not need special training and only needs to read the category descriptions to make a judgment, so the clarity of the category descriptions directly determines classification accuracy.
50%Unverifiedn8n has hundreds of built-in integration nodes (Gmail, Slack, databases, HTTP requests, etc.) and also provides AI-related nodes such as Text Classifier and Agent.
50%