Creating an NLU model

How to create an NLU model

To create a new model, proceed as follows:

1. Navigate to NLU → NLU Models section.

2. Click Create in the right upper corner. The following dialog box opens:

3. Click Create New → Next. The following dialog box opens:

4. Fill in the fields below:

• Name: Model name.

• Group: A group of users who can access the model.

• Domain: Based on the selected domain, an out-of-the-box understanding of intents and entities will be available using Omilia’s pre-tuned xPacks. Regardless of the selected domain, you can add your own data to augment your model’s understanding using machine learning. Available domains are the following:  

  • The Universal domain offers out-of-the-box entities understanding only. In case you want to also include intents, you can build your own fully custom intent understanding.
    The Universal 2.0 domain offers out-of-the-box entities that are not linked to any specific domain. This means that you can use it for any domain you might want to work with and you want to take advantage of this domain’s entity variety.

    • Enable US Addresses add-on: This checkbox is available for the Universal 2.0 domain only. If marked, the US Addresses add-on will be activated for your model.

  • The Custom domain has no out-of-the-box understanding and you can fully customize it by adding your own data and uploading your own custom NLU Logic.

  • The Generic domain provides out-of-the-box intents that cover multiple domains. It comes pre-trained and can be deployed as is, allowing for immediate use without additional customization.

  • Banking 2.0

  • Car Retail

  • Energy

  • Telecommunications

• Language: The NLU model’s Language. Out-of-the-box understanding is for the language you select.

  • Enable Multiple Intent Recognition: This checkbox is available for English(US) and English(GB) models. If checked, it allows handling scenarios where a single user input might involve more than one intention. For more information about this feature refer to the Double intent recognition section.

• Training Set: Add a file with training data in TXT, CSV, or TSV. It allows you to add your own custom training data to augment the model’s intent understanding using machine learning. This is an optional field.

  • Click the Sample File button to download an exemplary training set.

• Description: Provide a short description of the NLU model. This field is optional.

5. Click Create to confirm. The model is created and shown in the NLU Models list.

The Machine learning software version of a created model is automatically set to the latest one. However, you can import pre-trained models of previous versions if needed.

A model that is not the latest version is indicated with an exclamation icon, as shown below:

Model drill-down page

After having created a model, you are forwarded to the model drill-down page. Depending on the model domain, the drill-down page may look different:

• For models with a specific domain, you will see the out-of-the-box intents and entities, which are loaded automatically, if available. The out-of-the-box elements are labeled as RB-XP as shown below:

• For a custom domain model, no out-of-the-box understanding is available. If you are creating a custom domain model and have not uploaded any custom data yet, the model will look as follows:

• If you have uploaded your custom data by using a TXT, CSV, or TSV file, the uploaded intents and utterances will also be visible. The custom elements are not labeled, as shown below:

The drill-down page header reveals the following information about the model.