GenAI Feature

The GenAI feature in Chat360 is designed to enhance the platform's capability to understand and respond to user intents and entities more accurately using advanced AI models. This feature leverages data-driven insights and provides powerful reporting capabilities to improve user interactions.


  • Key Components
    • Intent: Definition: An intent represents the purpose or goal behind a user's input. It helps the chatbot understand what the user wants to achieve through their message.

  • Example Intents:
    • hungry_user: The user indicates they are hungry.
    • name: The user provides or requests a name.
    • place: The user mentions or asks about a place.
    • book_test_drive: The user expresses the intention to book a test drive.
  • Usage in Bot Building:
    • Intent Recognition:
    • The bot uses Natural Language Processing (NLP) to analyse user inputs and match them with predefined intents.
    • Example: If the user types, "I want to book a test drive," the bot identifies this input with the book_test_drive intent.
    • Intent Handling:
    • Once an intent is recognized, the bot triggers a specific workflow or set of actions related to that intent.
    • Example: For the book_test_drive intent, the bot might proceed to ask the user for their preferred date and time for the test drive.
    • Intent Management:
    • In the GenAI Hub, users can create, update, and delete intents. Each intent can be linked to specific phrases or keywords that help in its recognition.


  • Entity
    • Definition: An entity represents a specific piece of information in a user's input that provides context to an intent. Entities are often nouns or data points that the bot needs to extract to understand and process the user's request.

  • Example Entities:
    • date: Specific dates mentioned by the user.
    • name: Names of people or objects.
    • location: Places or locations referred to by the user.
    • product: Specific products mentioned in the conversation.
  • Usage in Bot Building:
    • Entity Extraction:
    • The bot identifies and extracts entities from the user input to gain more context.
    • Example: In the input "I want to book a test drive for July 15," the bot extracts "July 15" as the date entity.
    • Entity Utilisation:
    • Extracted entities are used to personalise and contextualise the bot's responses.
    • Example: After recognizing the date entity, the bot can respond with, "You have booked a test drive on July 15."
    • Entity Management:
    • In the GenAI Hub, users can define and manage entities. Entities can be linked to specific intents to refine the bot's understanding and response logic.
    • Example: The book_test_drive intent might be linked with entities like date and location to ensure the bot gathers all necessary information for the booking process.

  • Training Documents
    • The Chat360 platform offers robust capabilities for training your data using various advanced AI models. Here's a detailed overview of the Training Documents section, focusing on the available options for training data:

  • Training Options
    • Global GPT-3.5:
    • Description: This option leverages the globally trained GPT-3.5 model. GPT-3.5 is a powerful language model developed by OpenAI that has been trained on a diverse range of internet text.
    • Benefits: High generalisation capabilities.
    • Extensive language understanding and generation capabilities.
    • Suitable for a wide variety of use cases without the need for domain-specific customization.
    • Modified GPT-3.5:
    • Description: This version of GPT-3.5 is customised or fine-tuned to better suit specific needs or domains. By adjusting the training data or parameters, the model can be tailored to perform better in particular contexts.
    • Benefits: Enhanced performance in specific domains or tasks.
    • Improved accuracy and relevance for domain-specific queries.
    • Flexibility to incorporate specific knowledge bases or terminologies.
    • Large Language Models (LLMs):
    • Description: Beyond GPT-3.5, the platform supports training with other large language models (LLMs). These models can be selected based on the specific requirements of the project.
    • Benefits: Flexibility to choose from various LLMs depending on the use case.
    • Ability to leverage the strengths of different models for specialised tasks.
    • Potential to use state-of-the-art models that offer the latest advancements in language processing.

  • Bot Workflow Integration
    • The GenAI feature integrates seamlessly with the bot-building interface. Users can create custom workflows to handle various user intents and entities.

    • Components: Various components like custom input, multiple choice buttons, date/appointment selectors, etc., can be used to build interactive and responsive bots.
    • Workflow Example: A workflow for booking a test drive involves capturing the user's intent, requesting a date, and confirming the booking. The nodes in the workflow include:
      • Start
      • Custom Input
      • Intent Recognition
      • Entity Extraction
      • Confirmation Message
  • Using the GenAI Feature
  • Step-by-Step Guide
    • Accessing GenAI Hub:
    • Navigate to the GenAI Hub through the sidebar on the Chat360 platform.
    • Access the Intent & Entity section to view and manage intents and entities.
    • Creating and Managing Intents:
    • Click on "Add Intent" to create a new intent.
    • Provide a name and description for the intent.
    • Update or delete existing intents using the action buttons next to each intent.
    • Training Documents:
    • Upload documents for training the AI model by clicking on "Train Document".
    • Select the document type (Website, PDF, etc.) and upload the relevant files.
    • Manage and view uploaded documents through the document management interface.
    • Building a Bot Workflow:
    • Navigate to the Bot Builder from the sidebar.
    • Use components like "Custom Input", "Intent", "Entity", and "Message" to design the bot workflow.
    • Link the components to define the interaction flow.
    • Save and publish the bot to make it live.
  • GenAI Component:
    • The GenAI component elevates the conversational AI experience by offering integration with different language models. Users can customise their chatbot’s response generation according to the desired complexity and style by selecting from the following model options:
    • LLM (Large Language Model): Utilise the robust capabilities of a general Large Language Model for a broad range of topics and conversational contexts. This option is suitable for diverse interaction scenarios, providing reliable and comprehensive responses.
    • GPT-3.5: Opt for GPT-3.5 to harness an advanced AI language model renowned for its nuanced understanding and generation of human-like text. This option is ideal for creating conversations that require a deep understanding of context and subtleties in various domains.
    • LLM in GPT-3.5: Combine the strengths of a general LLM with the specific advancements of GPT-3.5. This hybrid approach allows the bot to handle a wide array of queries while benefiting from the refined conversational abilities of GPT-3.5, particularly effective for specialised interactions.


  • Example Use Case
    • Booking a Test Drive:
    • The user initiates the interaction by indicating the desire to book a test drive.
    • The bot captures this intent using the book_test_drive intent.
    • The bot requests the user to provide a date for the test drive using a custom input component.
    • The provided date is captured as an entity.
    • The bot confirms the booking with a message incorporating the captured date entity.
  • Benefits of GenAI
    • Improved Accuracy: The AI model's training on relevant documents and specific intents and entities ensures more accurate responses.
    • Customizable Workflows: Users can build highly customised bot workflows to suit specific business needs.
    • Data-Driven Insights: The integration with the GenAI Knowledge Hub provides valuable insights and reporting capabilities to optimise user interactions continuously.


Up Next

WhatsApp Business API