IBM Watson® Assistant is a question-and-answer system that provides a dialog interaction between the conversation system and users. This style of interaction is commonly called a chatbot.
Chatbot projects that use Watson Assistant involve three phases: scope, design, and integrate.
In the scope phase, you gather requirements for the conversation and how customers support the use case today. They might have a script, coded procedures, or other artifacts.
You define personas, create an empathy map, and build a system context diagram. Then, you extract the potential list of intents. Intents are the purposes or goals that are expressed in a user’s input, such as answering a question or processing a payment. After you define intents, you assess the sentences that lead to those intents.
In the design phase, you create an instance of Watson Assistant and use its builder tool to define the intents and the entities. An entity represents a class of object or data type that is relevant to a user’s purpose. At the end of the design phase, you start the dialog flow and unit-test it.
Finally, in the integrate phase, you develop the web app or microservice that interacts with Watson Assistant. You implement the business logic to handle the conversation context, and add other components to complement the business requirements, such as the IBM Watson Retrieve and Rank Service, ODM business rules, or IBM BPM process.