Have not found an active data mining forum yet so asking here.
I'm new to data mining but since I have the most theoretical comp.
sci. background at my work I've been given the task of setting up a
data mining system. The problem is that our setup seems pretty non

Please comment on the approach illustrated in :
http://www.zhmicro.com/Database.pdf
On July 19 this was posted at comp.object.
Regards,

I miss Fabian Pascal's criticism of the database industry, and
contributions to clear thinking on the relational model, as
dbdebunk.com has been silent for some time now.  I noticed recently,
however, that he is posting  (though not writing) on his other passion

1)
The following quote ( well I shortened it a bit ) is from a chapter
briefly describing MARTIN E-R notation
"Say we have entities ORDER and PRODUCT. One ORDER must include at

We are currently in the process of attempting to come up with a
solution to redesign our database. I am not going to use real
examples, but I will attempt to give enough information to enable
somebody to help me.

1)
Before I ask the question I must point  out that I understand the
difference between logical and conceptual level. Thus, conceptual
model represents DB design independently of the underlying logical and

No big reason for wondering about this, the uses may seem obscure but
also seem interesting to me, David C and his notation thread has sent me
off on what could be another wild goose chase.  Gave this post a
different subject so as not to usurp his.

Using the notation [A B C]  for  <NOT> (A <AND> B <AND> C), etc.
The following [  A [B]]   means "A implies B"  for Boolean algebra.  What is
the corresponding thing for Relational Algebra?
Also,  I'm trying to come up with a bracket notation for a "literal

I just started learning about databases and already I'm totally lost
1) When you use,for example, SQL statements to create tables and
implement relationships with  primary and foreign keys, do you operate
at level three ( internal/logical model ) of ANSI/SPARC architecture,

I am looking for solution (technique) that will allow me, while using
traditional B+ tree, to obtain key from value without creating
"reversed" table. Some overhead is acceptable but not "reversed"
table. Looping is also not an option.

D&D algebra defines the OR operator as
R(x,y) <OR> S(y,z) = {(x,y,z) | (x,y) in R & z in Z} union {(x,y,z) |
(y,z) in S & x in X}
since application of both AND and OR operators monothonically

1)
I still don't understand whether term data model (or data model
instance) refers to mechanism ( implemented as software ) that defines
what kind of structures ( tables, relationships etc ) system

Which relational operation perform faster in general, JOINs or SELECT
subqueries?
Suppose we have these tables:
create table departments (

I'm having trouble with a query concept.
I know that:
select max(order_date) from orders;
will return the date of the newest order, and that:

tables:
1. publication(pubid, title)
2. book(pubid, date)
3. journal(pubid, date)