n8n-nodes-rckflr-ner

n8n-nodes-rckflr-ner

This is an n8n community node for Named Entity Recognition (NER). It allows you to extract structured information from unstructured text, identifying and categorizing entities like persons, organizations, locations, dates, and more. This node leverages state-of-the-art machine learning models from Hugging Face Transformers.js, running them locally within your n8n instance.

Languages: English | Español | Português

Features

• Local Processing: All computations run directly on the server hosting your n8n instance. No data is sent to third-party services, ensuring privacy and security.
• No External API Keys Needed: This node is completely free to use and requires no paid subscriptions or API keys.
• Reliable Model: Uses a powerful, English-language model (Xenova/bert-base-NER) to ensure stability and performance.
• Customizable Aggregation: Control how the model groups recognized token parts into a single, coherent entity.

Important Note: The performance and speed of this node depend heavily on the hardware resources (CPU, RAM) of the machine running n8n. The first time the node is used, the model will be downloaded and cached, which may take some time.

Installation

• Go to Settings > Community Nodes in your n8n instance.
• Select Install and enter n8n-nodes-rckflr-ner.
• Click Install to add the node to your n8n instance.

Usage

The node takes a string of text specified in the Input Text property and outputs a single item with a new field containing an array of the entities found.

Important: This node will always output exactly one item. It does not process multiple items from its input. The Input Text parameter is always required.

Node Properties:

• Input Text: The text you want to analyze.
• Aggregation Strategy: Defines how to group word pieces into a single entity (e.g., "New" and "York" becoming "New York"). The simple strategy is recommended for most cases.
• Output Field Name: The name of the field where the resulting array of entities will be stored.

Example Workflow

You can find an example workflow in the examples directory of the repository or by simply copying the code below and pasting it into your n8n canvas.

Coming soon

License

MIT