In an expert system, there are two basic techniques for information processing and issue solving: Forward Chaining and Backward Chaining.

 

Forward Chaining

Forward chaining is a reasoning strategy where the system starts with a set of known facts and applies rules to draw conclusions. It is a data-driven approach that works by continually adding new facts to the knowledge base as they become available, leading to the desired conclusion.

Example: In a medical expert system, the system might start with the known symptoms of a patient and apply a series of rules to diagnose the illness.

Backward Chaining

Backward chaining is a goal-driven reasoning strategy where the system starts with a goal or hypothesis and works backward to determine which facts support the conclusion. This approach is useful when the goal is known, but the evidence is uncertain.

Example: In troubleshooting a computer system, the expert system might begin with a known malfunction and work backward to find the root cause.