Create an NLP question to match a user's question against a phrase of intent. For example, "book appointment", "I need a doctor" and "when are you available" are all different ways of expressing the intent of booking an appointment.
The chatbot compares the user's question against different ways the same intent may be phrased, using an algorithm that accounts for slight differences in phrasing or word choice and misspelled words, so that the user's response can be accurately matched without predicting all possible keywords or exact phrases in advance. You can also create a rewrite utterance task to use an LLM model to rewrite questions that the chatbot has difficulty matching.
If your chatbot supports multiple languages, you can also specify the language of your training phrases. Language options are only displayed if your chatbot has multiple languages configured.
NLP questions can also use entities as placeholders that can match multiple values you define. Entities help streamline your training phrases to make them more manageable. For more information, see Entities.
The chatbot attempts to match a question by keyword before checking the NLP questions.
You can use NLP questions as:
- Inbound questions that match either in a specific conversation flow or any conversation flow.
- Outbound questions that only match after a specific passage.
Also see: