Learn About Designing Chatbots with Oracle Digital Assistant

Your customers expect to be able to get answers from an intelligent chatbot within their chat application, instead of opening another application or accessing a web browser. Oracle Digital Assistant provides a platform for the development of intelligent digital assistants and chatbots to meet your customer's needs and integrate with your back-end services.

Unlike a web or mobile application, a chatbot is designed to be conversational, using natural language.

Architecture

The Oracle Digital Assistant platform supports the development of digital assistants and individual skill chatbots.

The Oracle Digital Assistant platform includes the tools to create and configure the following components:

• Digital assistants: The central chatbots that coordinate the user interaction with one or more skill chatbots, by using intelligent routing.
• Skill chatbots: The individual chatbots that engage with users to complete a task, and use natural language processing to resolve user intent and extract information from the message sent through conversational channels.
• Channels: The connectors that bind a digital assistant or an individual skill chatbot to messaging service, such as Facebook, WeChat, mobile, web or Alexa. Chatbots can also be connected to a live agent, including Oracle Service Cloud and third-party applications.
• Skill Store and Skill Catalog: Chatbot developers and administrators can add pre-built skill chatbots to their digital assistants, including SaaS service skill chatbots for ERP, CX and HCM.

Description of the illustration digital-assistant-arch.png

The architecture diagram illustrates a typical flow, where a channel is connected to a digital assistant:

1. You configure which messaging applications (channels) your customers use to interact with your digital assistant.
2. Oracle Digital Assistant routes your customer's messages to the appropriate skill chatbot, based on your customer's initial conversational input (utterances).
3. The skill chatbot identifies a specific intent and collects data from the customer and invokes back end services or initiates a process (like a booking).
4. The service returns information to the component, which can be added to the chatbot response.
5. The skill chatbot returns the response through the digital assistant to the messaging channel.

About Oracle Digital Assistant

Oracle Digital Assistant provides the platform that you can use to build, test and deploy chatbots.

Your digital assistant is the central point of contact for all the conversational experiences you provide to your customers. A digital assistant can route conversations to one or more skill chatbots, covering a broad set of business domains from a single interface. A digital assistant coordinates the search for an appropriate chatbot to support a specific service.

With Oracle Digital Assistant, you can develop assistants and skill chatbots that benefit from a more natural conversational user interface, through text or speech, to your enterprise systems. By using artificial intelligence (AI) and machine learning, Oracle Digital Assistant can detect what your customer is trying to achieve (their intent) and respond appropriately. With custom components, you can collect data and results of transactions from API connections to your back-end enterprise applications and information sources. You can use the platform tools to build and train your digital assistant without the need for specialist AI skills. Your digital assistant can then be exposed through many chat and voice channels, a custom mobile app, or your website.

Oracle Digital Assistant provides the following key features to chatbot developers:

• Natural language processing: You can focus on building your skill chatbots with a declarative interface, and to set up the intents, entities and dialog flow without having to worry about the processing algorithms, understand the natural language, or classify these inputs.
• Conversational flow and context: Allows you to define the appropriate conversational flows and sub-flows, by using the AI powered intent detection that is combined with a powerful state machine to maintain the context of the conversation.
• Question and answer (QnA): Allows you to create QnA skill chatbots, one of the most common uses for a digital assistant, by importing a set of existing questions and answers from a spreadsheet. You can combine QnA and intents in the same conversational flow within a skill chatbot.
• Form-based data input: Allows you to create and incorporate forms into customer conversations for the times when a form is the best way to capture structured information, to let your digital assistant combine structured and unstructured interactions.
• Live agent handover: Allows you to configure Oracle Digital Assistant to move the conversation to a live agent in Oracle Service Cloud (RightNow). When the transfer occurs, the human agent can view the conversation history, so the customer does not need to repeat responses they provided during the chatbot session.
• Enterprise integration: Allows you to use custom components to create modules that can be invoked during conversations to enable customers to perform actions to resolve issues. For example, a module can retrieve information, or perform transactions through APIs to your applications. You can configure your digital assistant to complete any task that your available APIs allow it to perform.

Learn more about Oracle Digital Assistant by watching this video:

About Required Services and Roles

This solution requires an Oracle Digital Assistant account with a ServiceDeveloper role.

See Learn how to get Oracle Cloud services for Oracle Solutions to get the cloud services you need.