Chat Data icon

Chat Data

Basic Chat Data Node

Actions8

Chatbot → Retrain Chatbot

AI-generated

Overview

This node operation allows retraining a chatbot by providing new training data. Users can specify the chatbot by name or ID and supply custom source text, question and answer pairs, URLs to scrape for additional training content, and scraping options such as cookies, content extraction preferences, and CSS selectors to include or exclude. This operation is useful for updating and improving the chatbot's knowledge base with fresh or more relevant information.

Use Case Examples

  1. A user wants to update their chatbot with new FAQs by providing a list of question and answer pairs.
  2. A user wants to retrain the chatbot using content scraped from specific URLs, with options to exclude certain page elements.
  3. A user provides custom source text to enhance the chatbot's understanding of a particular topic.

Properties

Name Meaning
Chatbot Name or ID Identifier for the chatbot to retrain, selectable from a list or specified via expression.
Source Text Custom text input to train the chatbot with, enhancing its knowledge base.
Questions & Answers A collection of question and answer pairs to train the chatbot on specific topics.
URLs to Scrape List of URLs from which content will be scraped to train the chatbot.
Scraping Options Settings to control the scraping process, including cookies for authorization, whether to extract main content only, and CSS selectors to include or exclude specific elements.

Output

JSON

  • success - Indicates if the retraining request was successful.
  • chatbotId - The ID of the chatbot that was retrained.
  • status - Status message or result from the retraining operation.

Dependencies

  • An API key credential for authentication to the chat data API service.

Troubleshooting

  • Ensure the chatbot ID is valid and accessible with the provided API credentials.
  • Verify that the source text, Q&A pairs, or URLs are correctly formatted and not empty if provided.
  • Check network connectivity and API endpoint availability.
  • Common error messages include invalid chatbot ID, authorization failures, or malformed request data. These can be resolved by verifying credentials, input data, and API endpoint configuration.

Discussion