home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!stanford.edu!morrow.stanford.edu!sumex-aim!rice
- From: rice@sumex-aim.Stanford.EDU (James Rice)
- Newsgroups: comp.ai
- Subject: Re: AI and forecasting
- Followup-To: comp.ai
- Date: 7 Jan 93 09:27:28
- Organization: Knowledge Systems Lab, Stanford University
- Lines: 46
- Message-ID: <RICE.93Jan7092728@hpp-ipc-2.stanford.edu>
- References: <79388@hydra.gatech.EDU>
- NNTP-Posting-Host: hpp-ipc-2.stanford.edu
- In-reply-to: imfacks@prism.gatech.EDU's message of 7 Jan 93 16:53:50 GMT
-
- In article <79388@hydra.gatech.EDU> imfacks@prism.gatech.EDU (K.K. Srinivasan) writes:
-
-
- Hi:
- I am interested in the use of AI techniques (Knowledge Based Systems,
- Machine Learning and Neural Nets) for the purposes of forecasting. Can
- some one give me pointers / references in the above area?
- Thanks.
- KK
-
-
- Being partisan, I would highly recommend the book:
-
- "Genetic Programming: On the programming of computers by means of
- natural selection"
- By John R. Koza. MIT Press 1992 (just out)
- This has just been announced as the alternate selection of the
- Library of Science book club thingy, so if you get books from
- them it would be a cheap(er) way to pick up a copy.
-
- There's a movie that goes with the book:
-
- "Genetic Programming - the Movie"
- By John R. Koza and yours truly also MIT press 1992.
-
-
- Not knowing what sort of forcasting you mean (I'd guess financial time
- series, but you might mean weather), I'd also suggest trying to find
- out what Pelkey (sp?) and Farmer are up to, but I hear that they are
- actually making money and are unlikely to tell anyone how they do it.
- However you might get some pointers from the publications that come
- out of the Santafe Institute. Addison Wesley publishes a whole series
- of books/monographs from them.
-
- Last year's IJCNN proceedings are on a CD ROM, so you might be able
- to search through them easily for any useful references. The CD is
- only readable on PCs, though :-(.
-
- My guess is that anyone who publishes anything about time-series
- prediction has a good chance of being a loser. The winners keep
- quiet.
-
-
-
-
- Rice.
-