Ben:
> This kind of simple heuristic is not generally workable, though it may
> work OK in some circumstances. Of course, this kind of thing has been
> tried before many times by others over the last decades.
>
> IMO this kind of rule cannot effectively be programmed into a system,
> if one wants general functionality ... instead such mappings must be
> *learned* ... taht is the Novamente approach.
> work OK in some circumstances. Of course, this kind of thing has been
> tried before many times by others over the last decades.
>
> IMO this kind of rule cannot effectively be programmed into a system,
> if one wants general functionality ... instead such mappings must be
> *learned* ... taht is the Novamente approach.
I am also saying that the mapping must be learned. In addition, the correct mapping to be learned would be the very rule that I stated, which is the most general rule. Thus, the learning method should be ILP (inductive logic programming).
If you use a numerical mapping method, it should be equivalent to the rule I stated. But most likely using the numerical method may end up with a more erratic mapping.
I made a slight mistake in the rule earlier, it should be:
IF
n is significantly > the average / usual number of a thing X
n is significantly > the average / usual number of a thing X
or n is significantly > half the elements in a set Y
THEN
"there are many Xs"
or "there are many in Y"
YKY
THEN
"there are many Xs"
or "there are many in Y"
YKY
To unsubscribe, change your address, or temporarily deactivate your subscription, please go to http://v2.listbox.com/member/[EMAIL PROTECTED]
