It seems to me that this wouldn't work. There are not so many different types of phrasings, so it would be fairly simple to write a parser generator which would then pass to a very basic interpreter to solve them.
For example, to solve the "Which of these is a T: W, X, Y or Z?" you would just put in a rule like "BodyPart ::= Foot | Knee | Leg | ..." "DayOfWeek ::= Monday | Tuesday | ..." "Color ::= Red | Blue | ..." and then have it match against those.
Maybe the next time I have some free time I'll see if I can go and implement it.
For example, to solve the "Which of these is a T: W, X, Y or Z?" you would just put in a rule like "BodyPart ::= Foot | Knee | Leg | ..." "DayOfWeek ::= Monday | Tuesday | ..." "Color ::= Red | Blue | ..." and then have it match against those.
Maybe the next time I have some free time I'll see if I can go and implement it.