Velocity Reviews - Computer Hardware Reviews

Velocity Reviews > Newsgroups > Computing > NZ Computing > Data Mining

Reply
Thread Tools

Data Mining

 
 
Shane
Guest
Posts: n/a
 
      06-27-2007
Has anyone here dabbled in said black art?
What tools did you find helpful/worthwhile?
Any hidden catches you care to mention?

TIA
--
It is only two weeks into the term that, in a calculus class, a student
raises his hand and asks: "Will we ever need this stuff in real life?"
The professor gently smiles at him and says: "Of course not - if your real
life will consist of flipping hamburgers at MacDonald's!"

 
Reply With Quote
 
 
 
 
thingy
Guest
Posts: n/a
 
      06-27-2007
Shane wrote:
> Has anyone here dabbled in said black art?
> What tools did you find helpful/worthwhile?
> Any hidden catches you care to mention?
>
> TIA


Catches,

Normalise your data....if you only have one data source then it is
probably no huge issue...ie that source and that system's owner owns the
data and its syntax...

As an example of problems when you want to data mine and your data is
in-consistant...

eg Location WGTN, WN, wn, Wn, Wellington, wellington....

Who owns the data and is therefore in control ie facilities would
probably own the extension number a user is on, so they determine its
layout eg 5436 or ext5436 etc etc.

Now you might have marketing, sales, finance all having their own
conflicting data and [web]forms for inputting it with in-adequate checks
on the syntax/layout....

So when you go to data mine info,and say dump it into a mailing list or
phone list and its automated, mayhem results....

So if you have multiple owners of each field you have to determine who
is the owner and where you get that info from, eg a HR database might
hold the employee number, facilities databse the ext number....IT
database/AD the email address.....all on separate systems....

So datamining can be very useful...however the costs can be horrendious....

These days a green field site should consider this before they decide
much else, it is that fundimental...

This is why you get an enterprise architect...and no I dont mean a
consultant....most are useless...

regards

Thing










 
Reply With Quote
 
 
 
 
Shane
Guest
Posts: n/a
 
      06-28-2007
thingy wrote:

> Shane wrote:
>> Has anyone here dabbled in said black art?
>> What tools did you find helpful/worthwhile?
>> Any hidden catches you care to mention?
>>
>> TIA

>
> Catches,
>
> Normalise your data....if you only have one data source then it is
> probably no huge issue...ie that source and that system's owner owns the
> data and its syntax...
>
> As an example of problems when you want to data mine and your data is
> in-consistant...
>
> eg Location WGTN, WN, wn, Wn, Wellington, wellington....
>
> Who owns the data and is therefore in control ie facilities would
> probably own the extension number a user is on, so they determine its
> layout eg 5436 or ext5436 etc etc.
>
> Now you might have marketing, sales, finance all having their own
> conflicting data and [web]forms for inputting it with in-adequate checks
> on the syntax/layout....
>
> So when you go to data mine info,and say dump it into a mailing list or
> phone list and its automated, mayhem results....
>
> So if you have multiple owners of each field you have to determine who
> is the owner and where you get that info from, eg a HR database might
> hold the employee number, facilities databse the ext number....IT
> database/AD the email address.....all on separate systems....
>
> So datamining can be very useful...however the costs can be
> horrendious....
>
> These days a green field site should consider this before they decide
> much else, it is that fundimental...
>
> This is why you get an enterprise architect...and no I dont mean a
> consultant....most are useless...
>
> regards
>
> Thing


Awesome, thanks thing. Im expecting my data to be even more varied than
that, so Ive definitely found a challenge
--
Q: What is often used by Canadians to help solve certain differential
equations?
A: the Lacrosse transform.

 
Reply With Quote
 
thingy
Guest
Posts: n/a
 
      06-28-2007
Shane wrote:
> thingy wrote:
>
>> Shane wrote:
>>> Has anyone here dabbled in said black art?
>>> What tools did you find helpful/worthwhile?
>>> Any hidden catches you care to mention?
>>>
>>> TIA

>> Catches,
>>
>> Normalise your data....if you only have one data source then it is
>> probably no huge issue...ie that source and that system's owner owns the
>> data and its syntax...
>>
>> As an example of problems when you want to data mine and your data is
>> in-consistant...
>>
>> eg Location WGTN, WN, wn, Wn, Wellington, wellington....
>>
>> Who owns the data and is therefore in control ie facilities would
>> probably own the extension number a user is on, so they determine its
>> layout eg 5436 or ext5436 etc etc.
>>
>> Now you might have marketing, sales, finance all having their own
>> conflicting data and [web]forms for inputting it with in-adequate checks
>> on the syntax/layout....
>>
>> So when you go to data mine info,and say dump it into a mailing list or
>> phone list and its automated, mayhem results....
>>
>> So if you have multiple owners of each field you have to determine who
>> is the owner and where you get that info from, eg a HR database might
>> hold the employee number, facilities databse the ext number....IT
>> database/AD the email address.....all on separate systems....
>>
>> So datamining can be very useful...however the costs can be
>> horrendious....
>>
>> These days a green field site should consider this before they decide
>> much else, it is that fundimental...
>>
>> This is why you get an enterprise architect...and no I dont mean a
>> consultant....most are useless...
>>
>> regards
>>
>> Thing

>
> Awesome, thanks thing. Im expecting my data to be even more varied than
> that, so Ive definitely found a challenge



I'd expect it to be....we have had various consultants come in claiming
to give us all this "business intelligence" and then run away....the
hardest part is normalising I suspect, after that we dont need
consultants....something like Politicians....

regards

Thing








 
Reply With Quote
 
 
 
Reply

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Clustering (Data Mining) framework for java tobi Java 1 06-02-2005 03:35 PM
Data Mining Source Code In VB.NET or C# http://www.visual-basic-data-mining.net/forum ASP .Net 0 09-08-2004 07:24 AM
WebSite Data Mining [G]rumpy [O]ld [D]uffer Computer Support 1 07-15-2004 06:44 PM
Data Mining Rajesh Birelly ASP .Net 0 06-01-2004 08:32 AM
Data Mining Web Pages nirajranka@rediffmail.com ASP .Net 0 06-01-2004 07:15 AM



Advertisments