Using Ziya NLU

Follow the steps below to start using Ziya NLU:

1. Sign up for a Ziya NLU account at https://ziawin.com/nlu.
2. After signing up, you will need to create a new project and provide the necessary details.
3. Train the model by providing sample user messages and expected bot responses. You can do this through the Ziya NLU interface.
4. Test the trained model to ensure it is functioning correctly.
5. Integrate the trained model into your application or website. You will need to use the provided API endpoints to receive and respond to user messages.

If you have any questions or issues, please feel free to contact us through our support page on https://ziawin.com/nlu.

Your bot response can include a variety of elements such as text, images, videos, attachments, buttons, dropdowns, entities, and even Google Maps:

1. Text: You can simply send text messages to the user. For example, "Hello, how can I assist you today?"
2. Images and videos: You can send images and videos to the user by including the respective URLs in your message.
3. Attachments: To send an attachment, choose the "Attachment" option and select the file type (e.g. PDF, MP4, MP3).
4. Buttons: To include a button, provide a button title and a payload. When the user clicks the button, the bot can respond accordingly. For example, "Click here to view our product catalog" could be a button title, and the payload could be a link to the product catalog.
5. Dropdowns: A dropdown allows the user to select an option from a list. To create a dropdown, include the dropdown options and their respective payloads in your message.
6. Entities: Entities are predefined values that the bot can recognize and respond to. You can create entities for things like dates, times, and locations.
7. Google Maps: To include a Google Map in your message, include an iframe with the map's source URL.
8. Cards: You can use cards to display information in a visually appealing way. Each card can include an image, title, description, and buttons.
9. Slides: Slides allow you to present multiple cards or images in a carousel format, enabling the user to swipe or navigate through them.
10. External program codes using API: You can integrate external program codes using APIs to perform specific functions or retrieve data from external systems.
11. ChatGPT: ChatGPT is an advanced language model that can engage in dynamic and interactive conversations with users.
12. PDF Q&A: You can provide PDF documents with question and answer sections, allowing users to access specific information within the document.
13. Many more: The platform supports a wide range of features and capabilities. Explore the documentation or resources for a comprehensive list of supported functionalities.

To train your chatbot, you need to provide user messages and bot responses. You can also add buttons, slots, and attachments to enhance the conversation. Once you have provided the necessary data, you can train your model by clicking on the "Train" button. Our platform will use the provided data to train the model and optimize its performance. You can refine the bot's responses based on the training results, and improve its ability to engage with users.

Once you have input your user messages and bot responses and trained the model, you can test the chatbot by using the "Bot Test" feature on our platform. This allows you to interact with the chatbot and ensure that it is responding appropriately to user messages. You can also refine your bot's responses based on the results of the test.

Want to know more about web integration? Visit our blog:

Web Integration Blog by ziawin

Interested in Telegram integration? Learn more from our blog:

Telegram Integration Blog by ziawin

In Ziya, you can integrate APIs for slots to customize the response once the slots are filled. By default, Ziya will give a "Thank you for sharing the details" response, but you can also use an external API. Ziya supports both POST and GET methods. Here's an example of a GET method:

                  def example():
                    # Perform necessary operations and retrieve response
                    response = "Email sent successfully"
          
                    # Return success response
                    response_data = {"text": response}
                    return JsonResponse(response_data)
                

This is just an example. In the Ziya platform, you can configure the API by navigating to the "Add Slots" section. There, you can specify the API details once all the slots are filled. The API can be triggered automatically when the conditions are met.

For support, you can reach us at [email protected].