Database

[keys]

wot is databasewarehousing? just a brief description will be great. I know data mining is linked with it.

[Stevee]

Its basicaly a repository for all data generated whithin a system, that can then be accessed for reporting by an outside reporting package.

It is usefull if a number of departments use the same system and management wants centralised reports.

It is also handy for generating different reports that are not imbedded in the system.

[keys]

thanks do u know how data mining is linked to db warehousing?

[OBonio]

I think data mining is the process involved in getting the data out that you want from the 'data warehouse'.

[vehelpfulguy]

Here is a web site about Data Mining.

http://www.the-data-mine.com/

I think data mining is more about the DISCOVERY of information in your database that you did not know was there.

For example, suppose you are a company that sells insurance, and you have a database with details of everyone you sell insurance to.

Now it is very easy to find out how many men insure with you, and how many women and so on.

But there may be other pieces of information in the database that would be useful, if only you knew what it was.

So Data Mining tools trawl through the database trying to find links or trends that may not be obvious.

So after a whle you may find that for example, 80% or the women who insure with you drive a Volkwagon Golf and also insure a pet with you.

This may help you in your future marketing or selling campaigns.

There are all sorts of uses of data mining.

[vehelpfulguy]

Data Warehousing is the collecting of data within a company or organization to help them plan or run the company.

The "opposite" of a Data Warehouse is a transaction database.

Lets take Amazon as an example. They sell books and all sorts of other things over the internet. They probably use a transaction database to take orders and track them.

So if they looked at the database today they may say we have �10 million pounds of orders waiting to be delivered.

But this database is ever changing, and if you looked at it tomorrow it may say you have �12 million pounds of orders waiting to be delivered. so their backlog is getting worse.

But because this database is ever changing it is difficult to keep track of things like the history of backlog of orders.

So what you do is every night you take a "snapshot" of part of your database and put it in another database called a data warehouse.

The historical data in a data warehouse rarely changes.

So if you take this snapshot every day then over time you can look into your data warehouse at this historical data and see if your company is performing better or worse.

So if the management see that the backlog of orders is getting worse and worse then they know they have to hire more people or improve their delivery process somehow.

This is just one example, there are lots of others.