I'm studying backwards chaining for an exam in AI and I understand the overall idea. However, I'm confused about the following slides from one of the lectures:
where $KB = \text{knowledge base}$.
My question: why is this version flawed? I get that in the flawed version when determining the rule $P \text{ if } Q_1$ we return something no matter what, but I don't see why its wrong to do that in this specific case. What am I missing?
Note: I didn't really know where to ask this question, so please feel free to guide me to another site if the question isn't relevant in this forum. Thanks in advance.
When solving a subgoal returns a failure, the second algorithm returns failure, while the first algorithm tries to find another applicable rule in the knowledge base. So the second algorithm can get stuck in a blind alley.