Skip to main content

Knowledge Base Overview

The Knowledge Base section allows users to create and manage collections of knowledge that can be used for AI and chatbot functionalities. These collections can be created from files, web pages, databases, or API configurations.

Key Features:​

  • File-based Collections: Upload documents in formats like .txt, .json, .csv, and create a searchable knowledge base.
  • Web-based Collections: Enter URLs of web pages to gather information and form a knowledge base.
  • Database Integration: Connect to your databases and create collections of knowledge directly from database queries.
  • API Configuration: Pull data from various APIs by configuring endpoints, and update them periodically.

How to Use:​

  1. Browse Existing Collections: View and manage previously created knowledge base collections from the list.
  2. Create New Collection: Click on the New Knowledgebase button to start building a new collection.
  3. Add Knowledge Sources: Choose between files, web, databases, or API configurations as your data source.

Steps to Create a New Knowledge Base:​

The New Knowledge Base page allows users to create new collections of information that will power AI models and assist in automated tasks. Users can collect data from multiple sources, including files, websites, databases, or APIs.

Steps:​

  1. Enter Collection Name: Start by entering a meaningful name for your knowledge base collection. This name will help you identify it later when managing multiple collections.

  2. File Upload:

    • Choose the File tab.
    • Click on Upload Document to upload your document from the local system. Supported formats include .txt, .csv, .json, and more.

  3. Web Scraping:

    • Select the Web tab.
    • Enter the URL of the website you wish to extract data from.
    • Click on the + button to add it to the collection.

  4. Database Integration:

    • Choose the Database tab.
    • Connect to your database by entering the query and setting up a refresh interval.
    • The data will be fetched based on the given query.

  5. API Configuration:

    • Select the API Config tab.
    • Configure the API endpoint, and set up a refresh interval to pull data periodically.

  6. Save the Collection: After adding your sources, click Save to complete the process.

Example:​

The screenshot below illustrates the Knowledge Base Home page where no collections have been created yet, hence the No Data placeholder is shown. Once collections are created, they will be listed here for further actions.

The screenshot below also illustrates the New Knowledge Base page where users can upload files, connect to web pages, databases, or APIs, and form a collection for the knowledge base.

Key Benefits:​

  • Model Customization: Fine-tune models based on your specific data to significantly improve the accuracy and relevance of generated responses.
  • Flexible Dataset Management: Easily upload new datasets or reuse existing ones from the dataset library.
  • Full Hyperparameter Control: Adjust critical settings like batch size, learning rate, and epochs to optimize performance.
tip

Using multiple data sources like API, files, and databases can make your knowledge base richer and more versatile for various use cases.