Training

Only the Producer and the Developer roles are allowed to manage intents.

Intents

When automating human conversation, understanding of the user's intent is key. Giving the right answer to the wrong question is biggest mistake made by bots today, hence identifying the intention of the user message is the biggest challenge.

The DialoX platform offers a special section to manage all the intents that are relevant for your bot. Intents can be organized in multiple files. However, most bots only offer one file called "Intents".

The system differentiates between global and local intents. The local intents can be recognized by the circle icon in front of the row, see billing, sales and support in the above example. However, local intents are currently only available to selected customers by means of a feature flag.

Local intents are introduced to make their recognition within a specific context much better while needing less training phrases. This is possible due to the use of the LLM AI technology called chat GPT.

Add intents

You can add a new intent by clicking on the Add intent button. Now you can specify the label and description of the intent and start adding training phrases.

Enable 'only use locally' when you are using the intent only to trigger the next link in an Ask or Prompt node.

Disable 'only use locally' when you are using the intent to trigger outside an ASK or PROMPT node. These only make sense for usage in Intent triggers. Because these intents can be triggered without knowing the exact context upfront, these intents need more careful training. The following hints can give some guidance:

• Using complete sentences works better than single words (unless intents are only used for topic classification, as in the example above)

• Try to have roughly the same amount of phrases per intent

• The optimum amount of phrases per intent is roughly 20

• Try to reuse important words in multiple sentences

Local intents

Toggle Only use locally when you need to recognize a user intent within the context of a specific dialog or flow. If you want this intent to be able to be triggered anywhere in your bot you need to keep it global (so toggled off).

When you make sure the description is clear, you will only need a couple of training phrases (examples). For clear intents one might already be enough, like in the example above.

Local intents are introduced to make their recognition within a specific context much better while needing less training phrases. This is possible due to the use of the LLM AI technology called chat GPT.

Test intents

Intent training can be quite a tedious process. The AI will use its algorithm to calculate which intent to classify for any given user utterance. The way that is done by the AI remains a black box. This often requires quite some tweaking of training sentences before the required behaviour is achieved.

It is possible to test the classification of intents by running your bot and see if it behaves as expected. Another, quicker way is to test the intent classification within the training section of the platform.

By typing a sentence (user utterance) in the Sentence field in the top of the side panel on the right, the system will automatically run it against the AI.

In the above example we are testing the global intent for opening hours.

For local intents, you need to specify the context by adding the intents you want to test the utterance against in the Context field below. This field is however only available behind a feature flag so it can be tested by a selected group of customers first.