Re: DM: Book on datamining

[Andrew A. Kramer]

Well I'll throw my 5 cents in. There are a plethora of books on neural
networks, some very theoretical. Most of the theoretical books are 
written
from an engineering viewpoint rather than a statistical/data analysis 
one.
Thus I've yet to find what I would consider a great textbook on NNs.

However, there are a few books that are worth considering:

1. Joseph Bigus' book is a good introduction for business users who
    wish to know more about NNs.
2. Jeanette Lawrence's into to NN's is a good primer for people with
   some math background.
3. Christopher Bishop's book on Pattern Recognition and 
Classification is
excellent,
although one should have a good knowledge of NN basics as well as a 
strong math
background.

Andrew A. Kramer, Ph.D.
Future Analytics, Inc.

Warren Sarle wrote:

> Ismail Parsa writes:
> > ...
> > Among related subject matter areas, Swingler's book (Applying 
>Neural
> > Networks) does an excellent job of covering all aspects of NN 
>modeling.
>
> I beg to differ. Swingler's book contains numerous gross errors and
> has thereby received the rare distinction of being placed on "The
> Worst" list of books in the Neural Net FAQ. See:
>
> ftp://ftp.sas.com/pub/neural/FAQ4.html#A_books_worst_swingler
> ftp://ftp.sas.com/pub/neural/FAQ4.html#A_books_code_swingler
>
> --
>
> Warren S. Sarle       SAS Institute Inc.   The opinions expressed 
>here
> saswss@unx.sas.com    SAS Campus Drive     are mine and not 
>necessarily
> (919) 677-8000        Cary, NC 27513, USA  those of SAS Institute.