Qbasicnews.com

Pages: 1 2

Here is program output that I made for my AI class. It is called the "Calcugator" It's supposed to think like a human to solve arithmetic problems. Do you think this could have any application in the education programs industry?

Code:
(show.reduction "5/2 + 7 - (9 - (6 - 9 * 5 + 2 * 45)) + 44 * 99 - (3 / 7 + 999)" 1)

5/2 + 7 - (9 - (6 - 9 * 5 + 2 * 45)) + 44 * 99 - (3 / 7 + 999)

5/2 + 7 - (9 - (6 - 9 * 5 + 2 * 9 * 5)) + 44 * 99 - (3 / 7 + 999)

5/2 + 7 - (9 - (6 - 9 * 5 + 2 * 9 * 5)) + 44 * (90 + 9) - (3 / 7 + 999)

5/2 + 7 - (9 - (3 * 2 - 9 * 5 + 2 * 9 * 5)) + 44 * (90 + 9) - (3 / 7 + 99)

5/2 + 7 - (9 - (6 - 9 * 5 + 2 * 9 * 5)) + 44 * (90 + 9) - (3 / 7 + 999)

5/2 + 7 - (9 - (6 - 45 + 2 * 9 * 5)) + 44 * (90 + 9) - (3 / 7 + 999)

5/2 + 7 - (9 - (6 - 45 + 2 * 45)) + 44 * (90 + 9) - (3 / 7 + 999)

5/2 + 7 - (9 - (6 - 45 + 90)) + 44 * (90 + 9) - (3 / 7 + 999)

5/2 + 7 - (9 - (6 - 45 + 90)) + 44 * (90 + 9) - (3/7 + 999)

5/2 + 7 - (9 - (-39 + 90)) + 44 * (90 + 9) - (3/7 + 999)

5/2 + 7 - (9 - (-39 + 90)) + 44 * 99 - (3/7 + 999)

5/2 + 7 - (9 - 51) + 44 * 99 - (3/7 + 999)

5/2 + 7 - -42 + 44 * 99 - (3/7 + 999)

5/2 + 7 - -42 + 44 * 99 - 6996/7

5/2 + 49 + 44 * 99 - 6996/7

5/2 + 49 + 4356 - 6996/7

5/2 + 4405 - 6996/7

8815/2 - 6996/7

47713/14

Initially, I created a standard, linear way of solving problems. Then I expanded on it to preferentially solve available subproblems (those that can be solved in any one stage according to the rules of mathematics) and expand out numbers based on computational cost criteria.

I will upload the source and documentation if anyone's interested. (LISP)

it's an expression evaluator that substitutes. i made one in qb once when i twas a wee lad. there are algebraic problem solvers aplenty however. a man in 1996 used an algebra program to do some proof that had stumped scientists for over 60 years. and that was 7 years ago, who knows where it's at now?

Quote:when i twas a wee lad.

Just wanted to say that "twas" means "it was" so... don't try to sound olde english Tongue

Yes, but it evaluates and substitutes based on a data table of strength values. And the further you get, the more inclined one is for evaluating instead of substituting.

Now, I am not trying to make and solve a proof or anything, just show stupid people how to do arithmetic...

Oh, and for that problem solver. The method is supposedly virtually discredited (so says my professor). The reason is that the person who made it nearly put in every bit of knoweledge that it put out. And any that don't just follow some stupid little logic system.

Not that mine doesn't, but it's not just cold logic.

Useful? Maybe. Profitable? You said profitable??? [Image: eek7.gif]

Quote:Useful? Maybe. Profitable? You said profitable???

Hehehe, ya.... I only know of n00bs trying to get money off of semi-useful all-in-all shitty programs.

Not saying your program sucks, I just mean... no one would really pay for it.

*shrug* i dont know. i've never made one. but substitution works for that stuff i guess. i also made a program that solves derivatives step by step (http://derivatives.netfirms.com) in a similar fashion, but trust me, there are far more capable programs out there than anything you or i probably gots (namely, http://mss.math.vanderbilt.edu/~pscrooke/toolkit.html)

oh, and @nikazu: f***e ye. Wink

I'll repost that link, 'cuz it's %&$&d up. http://derivatives.netfirms.com

Oh, and @ toonski: LOL.

It doesn't show intermediate steps. Show me something...with intermediate steps... that show how to solve the problem and make, say, 101 * 56 into (100 + 1) * 56........

Code:
f(x) = log(x^2)+x^3

f0(x) = x

f1(x) = f0^2

f2(x) = logf1+f0^3

(EDIT)

x is one of the answers that the other link gives, too.

If I had that link (http://mss.math.vanderbilt.edu/cgi-bin/M...vative.def) it would have saved some MAJOR time wastin' on horrific problems when I was doing Calculus a year ago.

Now that I've taken them, I don't have to take or apply Maths classes EVER again! YAY! 2+2=1!

2+2=5, dumbass...

(and my thingy doesnt do logs, so it doesnt mess wit' that shist)

Pages: 1 2

Agamemnus

toonski84

Ninkazu

Agamemnus

Hexadecimal Disaster

Ninkazu

toonski84

Hexadecimal Disaster

Agamemnus

toonski84