Articles in this section

How to train intents?

Intents are an essential component of building conversational interfaces that can interpret and respond to user input in a way that is natural and intuitive. That is why it is important to follow best practices to ensure that your conversational interface is accurate and effective.

Here are some best practices for training and maintaining intents:

  1. Start with a clear goal for each intent: Before you begin training your conversational interface, you should have a clear idea of what you want it to achieve. Each intent should serve a specific purpose, and the training data should be designed to help the model recognize and respond to that purpose.
  2. Identify common user phrases and variations that correspond to each intent: Once you have defined the goal for each intent, the next step is to identify the common phrases and variations that users might use to express that intent. By providing a diverse set of training phrases, you can help the model recognize variations in language and context.
  3. Use a diverse set of training phrases to ensure that the model can recognize variations in language and context: In addition to identifying common user phrases, you should also use a diverse set of training phrases. This will help the model recognize variations in language and context, which is essential for accuracy.
  4. Continuously test and evaluate the model's performance, and adjust the training data as needed: Even after your conversational interface is up and running, you should continue to test and evaluate its performance. This will help you identify any areas where the model is struggling and adjust the training data as needed.
  5. Regularly review and update the intent list to reflect changes in user behavior, new use cases, and evolving business needs: Finally, you should regularly review and update the intent list to reflect changes in user behavior, new use cases, and evolving business needs. This will ensure that your conversational interface remains accurate and effective over time.

By following these best practices, you can create a more accurate and effective conversational interface that can better understand and respond to user input.

 

Example intents in Customer Support

Here are three examples of intents in the customer support industry along with five training phrases for each intent:

Intent 1: Forgot Password

  • I forgot my password, what do I do?
  • Can you help me reset my password?
  • How do I recover my account password?
  • I'm not able to log in, what are my options?
  • Can you guide me through the process of resetting my password?

Intent 2: Order Status

  • Can you provide an update on my order status?
  • Where is my order currently?
  • When can I expect my order to be delivered?
  • I haven't received any updates on my order, can you help me?
  • How do I track my order?

Intent 3: Cancel Order

  • Can I cancel my order?
  • How do I cancel an order?
  • I need to cancel my order, what do I do?
  • Is it possible to cancel an order after it has been placed?
  • Can you guide me through the process of canceling my order?
Was this article helpful?
0 out of 0 found this helpful

More resources

  • Need support?

    Submit a request and our support team will assist you

  • Business hours

    Monday to Friday 8am - 5pm PST excluding US holidays

  • Contact us

    support@forethought.ai