Nautilus Systems, Inc. logo and menu bar Site Index Home
News Books
Button Bar Menu- Choices also at bottom of page About Nautilus Services Partners Case Studies Contact Us
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [Subscribe]

Re: DM: Software suggestion for clickstream data mining?


From: Gordon S. Linoff
Date: Mon, 07 Feb 2000 13:12:39 -0500
  • Organization: Data Miners


The process of analyzing clickstream data is really not different from
analyzing any other voluminous source of transaction data.  Generally, you
want
to extract features (such as paths) and use this for some sort of
optimization
(customer-focus, site-personalization, marketing, and so on).  Once featurs
are
extracted from the data, any of the usual suspects in data mining tools can
build models, because these features are just additional columns on use
profiles.

Primary Knowledge (http://www.primaryknowledge.com) is a company that I do
some
work with here in New York.  Their business is transforming clickstream data
into useful information at the customer and prospect level . . . currently
more
on the reporting side but with a keen interest in data mining.

Their business model is as an outsourced provider of services, so they do
the
processing and cleansing of clickstream data.  I realize that their web site
is
not that informative, so if someone would like to learn a bit more about
them
they can contact me (I am not an employee; they are a consulting client).

There are a number of other companies that are in or entering this market
place.  So, if you have a lot of money to spend and are looking for a
solution,
you may want to look at DataSage (which was recently purchased by Vignette),
Net Perceptions (which recently acquired KD1), Accrue (which recently
purchased
Neovista), Cogit, and Epiphany come to mind.  However, there are dozens of
companies in the clickstream analysis/e-crm/e-data-mining/etc. market.

Of course, the large web sites are doing it themselves.  Amazon, for
instance,
has developed in-house software for their recommendation engine and
clickstream
analysis.  Ditto for DoubleClick.  Yahoo purchased a company called
Hyperparallel in late 1998 and has a large group (dozens of analysts) doing
this type of work.

And, for dealing with the shear volumes of clickstream data, I would
recommend
software from companies such as Ab Initio or Torrent Systems, since they can
effectively handle the large amounts of data on scalable hardware systems.
Ab
Initio has little information on their web site (www.init.com), but has the
more impressive client list.  If you are serious and would like to learn
more,
I suggest reading Chapter 12 in "Mastering Data Mining:  The Art and Science
of
Customer Relationship Management" (Berry and Linoff; Wiley, 2000).  Or, if
you
are serious, you can contact me directly (once again, I do not work for Ab
Initio, but they are one of my consulting clients, both directly and
indirectly).

--gordon
Gordon S. Linoff, Founder
Data Miners, Inc.
gordon@data-miners.com

"Lynd D. Bacon" wrote:

> Great question!  Can respondents please post their responses to the
> list?  I'm sure many list subscribers would like to see them.
>
> Scott, would you summarize any direct replies you get for us and post
them?
>
> Thanks!
>
> Lynd Bacon
>
> At 05:52 AM 02/07/2000 -0800, Scott Nicholson wrote:
> >Does anyone have a suggestion for good software to aid
> >in clickstream analysis?  I'm looking for something
> >that goes beyond the typical analysis and provides
> >pathways of clicks, allowing the grouping of customers
> >by what they tend to look at.
> >
> >Thanks,
> >Scott Nicholson
> ...
> ---------------------------------
> LBA:Lynd Bacon & Associates, Ltd.
>         Homewood IL USA
>           www.lba.com
>         +1 708 957-0883
>         +1 708 957-1920 fax
> ---------------------------------





[ Home | About Nautilus | Case Studies | Partners | Contact Nautilus ]
[ Subscribe to Lists | Recommended Books ]

logo Copyright © 1999 Nautilus Systems, Inc. All Rights Reserved.
Email: firschng@nautilus-systems.com
Mail converted by MHonArc 2.2.0