Oracle data mining, not only good but affordable too!

Having been very impressed by what Microsoft has to offer I
wanted to learn more about what Oracle has to offer, and I am
equally impressed. I, like many others I suspect, tend to rather
overlook that Oracle does data mining; because Oracle has so much
on offer, some elements do tend to get neglected. In my case,
having been a Tru64 user for many years, Oracle was not an option
because they did not support the platform. But now that the mining
is integral to the kernel of the database it is universally
available, but old habits die hard and it has rather been off my
radar.

An aspect of the tool that I have to comment on is its pricing.
I hear a lot from Oracle users about how expensive everything is
getting, it therefore comes as a very pleasant surprise to find
that the Oracle data mining option is one of the great bargains
available today because it is affordable and when you look at what
you are getting it is a real Rolls Royce of capability and
features.

That Oracle data mining should be as sophisticated as it is
should not really be a surprise, because when Oracle went out to
acquire data mining capability they went after one of the top
offerings. The origins of the expertise comes from Thinking
Machines whose Darwin product was one of the best that I had seen;
but at the time I examined it for a leading Telco it was deemed by
my managers to be too expensive so we went for Clementine, but I
have always held in high esteem the experts behind Darwin. Having
been acquired by Oracle, those experts obviously not only had the
huge customer base to go after, but also the highly competent base
technology of the Oracle database itself. So they redesigned from
scratch and put the algorithms into the database to ensure that,
not only is the execution of the algorithms efficient, but the vast
amounts of data handling that typifies traditional data mining is
minimised. This cannot be emphasised enough because data mining
analysts traditionally only spend a tiny proportion of their time
actually modelling; something like 80% of effort is expended on
getting data to the tool to be used. So Oracle are leaving the
database in situ and mining it there, which saves a lot of effort
and will greatly increase productivity.

Whilst Microsoft may have impressed me by bringing data mining
to the masses, Oracle is pitching at a different segment. This is a
fully featured, highly sophisticated data mining capability to
enable professionals to operate against Oracle data sets with
productivity and precision. Oracle data mining has a broad range of
available algorithms, which enable it to undertake virtually every
kind of business and scientific analysis that one can think of. So
whether you want to run supervised learning to classify data and
make predictions, or you want to run unsupervised sessions to
discover anomalies and clusters, Oracle has the right tool for the
job.

When Oracle Data Mining in the database was first released I was
again discouraged from using it because the algorithms were there
but we had no commercially available means of easily accessing
them. SPSS have been making their front end tools available to
enable the benefits of the in-database offerings of Microsoft, IBM
and Oracle to be readily exploited, and now Oracle are developing
GUI based interfaces of their own, so again reasons not to examine
this product are becoming increasingly hard to find.

One of the features that Microsoft are offering next year is the
inclusion of data mining add-ins to Excel, but Oracle users need
not feel obliged to migrate to Microsoft in order to gain such
useful capabilities because Oracle themselves offer Excel users a
one click access to mine data whether it is held in Oracle or
Excel.

Oracle is where an awful lot of data exists today, and of the
general purpose databases (so I am excluding the likes of Alterian
from this) in my experience it is by far the most performant. With
the acquisition of Seibel and their BI suite Oracle created a
formidable BI offering, and now that I realise what a hidden gem
their data mining suite is, it really is a very hard act for others
to compete with. Whilst Microsoft is about bringing data mining to
the masses Oracle are giving the data mining professional a real
alternative to SAS and SPSS with an offering that is equally as
well featured, but which promises to outperform any standalone
offering.