chaining (1 Viewer)

[crammy90]

when a forward chaining inference engine cannot reach a conclusion, does it then begin to use backward chaining?

 

[VenomP]

Not 100% on this one, but I'm thinking that if the Expert System couldn't come to a conclusion as to what it was, it wouldn't be within its records.

I say this because there is no "fuzzy logic" in an expert system, only neural networks (where inference engines are not used).

Therefore, the wrong conclusion would be obtained from the database of results.

The human operating the system THEN could say "Well, this is not what I meant, I'll try backwards chaining". It's up to the operating human as to accept the result or try a different way.

 

[crammy90]

Not 100% on this one, but I'm thinking that if the Expert System couldn't come to a conclusion as to what it was, it wouldn't be within its records.

I say this because there is no "fuzzy logic" in an expert system, only neural networks (where inference engines are not used).

Therefore, the wrong conclusion would be obtained from the database of results.

The human operating the system THEN could say "Well, this is not what I meant, I'll try backwards chaining". It's up to the operating human as to accept the result or try a different way.

expert systems use fuzzy logic tho by using confidence and certainty factors to fire different rules :S
oh well lol