Skip to main content

Datasources

Incorporate the ability to choose from multiple data sources for selecting a knowledge base through diverse methods such as website scraping, uploading PDF, TXT, or CSV documents, or inputting specific information.

Upon initiating a website or document as a datasource, the system will commence content scraping, followed by training and storage of the extracted information in a vectorized database. The progress of scraping and training will be visually indicated through a progress bar until the process reaches completion.

Datasources

You will come across various statuses associated with each data source record, outlined as follows:

  • Pending: Indicates that the data source has been added, and content retrieval is pending.
  • In Progress: Signifies that content retrieval from the added data source is currently underway.
  • Synced/Success: Denotes successful content retrieval from the added data source.
  • Failed: Represents a failure in content retrieval from the source.
  • Sync Requested: Implies that you have initiated a request to re-sync the content from the source.
  • Training Requested: Indicates your request to train the model for the given data source.
  • Trained: Indicates the completion of training, allowing the AI Chatbot to utilize this content for responding to end users' messages.