Revolutionary AI Chatbot Platform: Kore.AI Bots
Table of Contents
- Introduction
- Getting Started with the Core Bot Builder
- Creating a Bot Configuration
- Creating a Standard Bot
- Creating a Multi-bot Configuration
- Understanding Task Types
- Dialogue Tasks
- Knowledge Tasks
- Alert Tasks
- Information Tasks
- Action Tasks
- Designing Dialogue Flow
- Using the Dialogue Builder
- Setting Node Properties
- Configuring Natural Language Training
- Providing Training Data
- Visualizing the Model
- Adjusting Engine Settings
- Enabling and Testing Channels
- Enabling Channels
- Configuring SDKs
- Testing Dialogues
- Publishing and Analyzing Performance
- Publishing Changes
- Analyzing Bot Performance
- Customizing Bot Settings
- Changing Bot Settings
- Inviting Collaborators
- Importing and Exporting Configuration
- Managing Bot Logs
- Monitoring Bot Dashboard
- Conclusion
Introduction
Welcome to an overview of the Core Platform and the Core Bot Builder. In this article, we will explore the features and functionalities of the Core Bot Builder, a web-Based IDE for designing, developing, and deploying bots on the Core Platform.
Getting Started with the Core Bot Builder
To access the Bot Builder, go to the Developer Core Portal and click on the Bot Builder icon located in the top right corner. The Bot Builder is a powerful tool that allows You to Create and manage bots on the Core Platform.
Creating a Bot Configuration
In the Bot Builder, you can create a new bot configuration or work with existing bots developed by you or shared with you by other developers.
Creating a Standard Bot
You can start by creating a brand new bot called a Standard Bot. This Type of bot is designed to perform specific tasks and functionalities related to a particular domain, such as banking, customer service, or e-commerce.
Creating a Multi-bot Configuration
Alternatively, you can create a Multi-bot Configuration, which allows you to combine multiple bots into a single configuration. This is useful when you want to create a bot that utilizes various bots for different functionalities, making it more versatile and powerful.
Understanding Task Types
In the Core Bot Builder, bots are composed of different task types, each serving a specific purpose. Let's explore these task types in Detail.
Dialogue Tasks
Dialogue tasks represent the functional behavior of the bots. For example, a banking bot may have tasks related to checking account balances, making payments, or processing transactions. The Dialogue Builder provides a visual drag-and-drop environment for designing the conversation flow between the bot and the user.
Knowledge Tasks
Knowledge tasks are used when you want the bot to be able to answer frequently asked questions. You can import a set of questions and train the bot to provide accurate responses without the need for backend lookups. The Core Platform supports importing knowledge from various sources and organizing it using ontology structures.
Alert Tasks
Alert tasks allow you to proactively engage the user in a conversation based on user-defined events or enterprise-defined alerts. This is useful when you want to send notifications or initiate conversations based on specific triggers, such as a new message or a system event.
Information Tasks
Information tasks enable the bot to retrieve information from backend systems based on user queries. It allows users to ask questions in natural language and receive Relevant information as a response. The platform supports data summarization and presenting the information using different UI elements within the conversation.
Action Tasks
Action tasks are used for performing specific actions, such as reporting a lost or stolen card. These tasks involve slot filling, where the bot may ask specific questions or prompt users to provide necessary information before executing the action. Once the required slots are filled, the bot performs the action, such as reporting a lost or stolen card.
Designing Dialogue Flow
The Dialogue Builder in the Core Bot Builder provides a complete visual drag-and-drop environment for designing the conversation flow between the bot and the user. Each node in the dialogue represents a step in the conversation, and you can set properties for each node to define its behavior.
Using the Dialogue Builder
To design the conversation flow, you can drag and drop nodes onto the canvas and connect them to create a sequence of steps. You can configure properties for each node, such as intents, entities, API calls, business logic, and validation. This allows you to define the behavior of the bot at each step of the dialogue.
Setting Node Properties
Each node in the dialogue has specific properties that can be configured. For example, intent nodes have properties related to intent recognition, entity nodes have properties related to entity detection, service nodes have properties for making API calls, and confirmation nodes allow you to validate user decisions before proceeding. You can also define messages to be displayed to the user at each step of the dialogue.
Configuring Natural Language Training
To accurately recognize user intents and entities, the Core Platform offers multiple natural language engines. You can provide training data to each engine to improve its accuracy and performance.
Providing Training Data
To train the machine learning models, you need to provide training data in the form of utterances for various intents. The Core Platform allows you to import utterances or manually define them. You can also Visualize the model to evaluate its performance and identify any training issues.
Visualizing the Model
The Core Platform provides a visual representation of the trained models, showing how well each utterance scores for a particular intent. This allows you to identify false positives and false negatives, enabling you to improve the training data and achieve better intent recognition.
Adjusting Engine Settings
The natural language engines in the Core Platform have adjustable thresholds and settings that can be modified to optimize their performance for your specific use case. You can fine-tune the parameters to achieve the desired balance between accuracy and detection.
Enabling and Testing Channels
Once you have designed and trained your bot, you can enable it to work in various channels, such as web, mobile, SMS, or Facebook Messenger. The Core Bot Builder provides easy-to-use configuration options for enabling and testing channels.
Enabling Channels
To enable a Channel, you can select the desired channel in the Bot Builder and follow the configuration steps provided. This allows your bot to Interact with users through different channels, expanding its reach and accessibility.
Configuring SDKs
To have more granular control over the bot's behavior, you can configure SDKs in the Apps and Agents view. This allows you to introduce custom logic and execution paths that Align with the specific requirements of the enterprise or application.
Testing Dialogues
The Core Bot Builder allows you to test the dialogues you have created directly within the platform. You can invoke the "Talk to the Bot" capability to simulate user interactions and monitor the conversation using the debug panel. This helps you ensure that the dialogue flows smoothly and the bot correctly understands user intents and responds accordingly.
Publishing and Analyzing Performance
Once you are satisfied with your bot's configuration and testing, you can publish the changes to make them live. This process involves approval from an administrator before the new configuration is pushed into production.
Publishing Changes
To publish your bot configuration, you can submit it for review and approval. Once approved, the changes will be deployed and made available to users.
Analyzing Bot Performance
The Core Bot Builder provides performance analysis tools to monitor and evaluate your bot's performance. You can track the number of interactions, analyze the accuracy of intent recognition, and identify popular tasks and channels. This helps you measure and optimize your bot's effectiveness and user satisfaction.
Customizing Bot Settings
In the Bot Builder, you have the flexibility to customize various settings and options for your bot. This allows you to tailor the bot's behavior and appearance to meet specific requirements.
Changing Bot Settings
You can change the name, logo, welcome message, and other settings of your bot. This allows you to personalize the bot and align it with your brand or application.
Inviting Collaborators
You can invite other developers to collaborate on the bot project. This enables multi-developer collaboration, allowing you to leverage the expertise and skills of multiple team members.
Importing and Exporting Configuration
The Core Bot Builder provides the ability to import and export bot configurations. This allows you to easily move configurations between environments or store them in a source control repository.
Managing Bot Logs
The platform maintains logs of all changes made to your bot configuration. This provides a history of modifications and allows you to review and revert changes if needed.
Monitoring Bot Dashboard
The Bot Dashboard provides a comprehensive view of your bot's performance. It displays metrics such as interaction volume, peak usage hours, intent accuracy, and popular channels. This helps you monitor the bot's overall health and make data-driven decisions to optimize its performance.
Conclusion
The Core Bot Builder offers a complete end-to-end solution for designing, developing, and deploying bots on the Core Platform. With its intuitive interface and powerful features, you can create highly interactive and intelligent bots that provide a seamless user experience. Take AdVantage of the capabilities of the Bot Builder to build and deploy your bots with ease and efficiency.