An expert system consists of several cooperating components: a knowledge base that stores facts and rules, an inference engine that applies those rules to reach conclusions, and a user interface through which a person communicates with the system. The user interface is the intermediary between the human operator and the inference engine. It accepts questions, data, and commands from the operator, converts them into a form the engine can process, and then presents the engine's conclusions, and often the supporting explanation, in terms the operator understands. Without this layer, a non-specialist could not draw on the encoded expertise.
The knowledge base does not mediate between anyone. Its function is storage and organization of domain knowledge acquired from experts, and it is consulted by the inference engine rather than by the operator directly.
A chatbot is a particular style of conversational front end. Such a program can serve as the interface for some systems, but it is an implementation choice, not a named structural component of the classical expert system architecture. Many expert systems use forms, menus, or command screens instead.
Since the question asks for the architectural component that connects the operator with the inference engine, the user interface is the precise answer.
Submit