Backward chaining is an inference method widely used in artificial intelligence, automated theorem provers and proof assistants. Backward chaining methodology can be described as working back from a goal. Many programming languages support backward chaining within their inference engines.
Backward chaining is also referred to as backward reasoning.
Backward chaining is used in logic programming through selective linear definite clause resolution. It's one of the most commonly used methods of reasoning with interference rules and logical implication. Backward chaining works backwards from a list of goals to determine whether there is any data to support them. An inference engine that uses backward chaining searches the inference rules until it finds a rule with a consequent that matches the desired goal. If the antecedent of this rule is not known to be true, the rule is added to the list of goals so that data can be found to confirm it.
0 Comments