PDA

View Full Version : AI and diet calculations?


Welcome to the Active Low-Carber Forums

Support for Atkins diet, Protein Power, Neanderthin (Paleo Diet), CAD/CALP, Dr. Bernstein Diabetes Solution and any other healthy low-carb diet or plan, all are welcome in our lowcarb community. Forget starvation and fad diets -- join the healthy eating crowd! You may register by clicking here, it's free!



Vp
Tue, Jun-25-02, 20:57
Besides linear programming, is it possible for the pc to
"understand" the person's nutritional desires and according to
them and the nutrient calculations to perform a diet creation?
Any other math (ideas) suggestions (besides linear programming
and AI) for diet calculations?
--

--------------------------------------------------------
----------
Όταν οι πύλες της ενόρασης ανοίξουν τότε τα πράγματα θα φανούν
όπως πραγματικά είναι..... Απεριόριστα

Wuzzy
Wed, Jun-26-02, 13:56
"VP" <bhvip@hotmail.com> wrote in message
news:<afal6m$3ev$1@nic.grnet.gr>...
> Besides linear programming, is it possible for the pc to
> "understand" the person's nutritional desires and according
> to them and the nutrient calculations to perform a diet
> creation? Any other math (ideas) suggestions (besides linear
> programming and AI) for diet calculations?

Thats a good idea, I read somewhere that the "paperclip" in MS
Office uses bayesian probability as a method of artificial
intelligence. (bayesian just means using prior probabilities
to come up with new probabilities) So you probably would have
to punch in daily intake and it would record this daily and
figure out probabilities of future diet as well as stuff you
might want to select to improve diet..

I would be interested in reading more about the MS OFfice
method as well,(not that it works) I totally lost the
reference.. Scientific American? probably not..

Also search google groups for AI and fuzzy logic..

Wuzzy
Wed, Jun-26-02, 13:56
I answer my own question:

I found the article I referred to regarding the paperclip it
was Science magazine.. It links the following
research.microsoft.com/~horvitz/lum.htm

Click on the PDFs for Conventions that took place on bayesian
inference.. a good review of the excel paperclip is on
ftp://ftp.research.microsoft.com/pub/ejh/lum.pdf

also in plain language is the following article: AI Magazine,
Summer 1999 v20 i2 p11 An Overview of Some Recent Developments
in Bayesian Problem-Solving Techniques. Peter Haddawy.

Ugh, I am constantly bothered by it. (joking)

Actually it is pretty cool, if you don't mind being babied..
(not for me) Maybe when computers get alot smarter..

Wuzzy
Wed, Jun-26-02, 20:56
MYPCOS@HOTMAIL.COM (wuzzy) wrote in message
news:<38bdd5da.0206260755.1afcab9e@posting.google.com>...
> I answer my own question:
>
> I found the article I referred to regarding the paperclip it
> was Science magazine.. It links the following
> research.microsoft.com/~horvitz/lum.htm

of note to doctors on the list is OVID's webweaver, you need
to subscribe to medline though to access the service, but it
is an artifical intelligence system you input the systems (in
plain english), the age of the patient how long they have the
symptoms and then it gives you the probability of each
disease...

very cool..

also after giving you a probability it gives an explanation as
well as links..