Question about keyword outrigger

View previous topic View next topic Go down

Question about keyword outrigger

Post  Booma on Wed Apr 23, 2014 2:56 am

In the Kimball book for dimensional modelling, there is a outrigger type dimension detailed. I have something similair in my datawarehouse, where a customer can have alot of keywords attached to him or her. However there are alot of different keywords. Like a 100. If I would follow the approach in the book, that would be a keyword outrigger of 100! (100x99x98x97...) Those are all the combinations of keywords. Granted that not all combinations will occur, but alot of them will.
I thought of skipping the bridge table, and make a dimension like this
customer_keyword_dimension
customer idkeyword
1C++
1C#
1Python
2Python
3C#
3Python
4C++
4SQL
4C#

So I'm like skipping the bridge table. But this means that this dimension will be n times as large as large as the customer dimension, where n is the average amount of keywords (probably 30). This is still much smaller than the bridge table approach, but still pretty large.

Is this a good approach or are there better alternatives? Would love to hear your insights
avatar
Booma

Posts : 12
Join date : 2014-03-10

View user profile

Back to top Go down

Re: Question about keyword outrigger

Post  nick_white on Thu Apr 24, 2014 5:09 am

My understanding of outrigger dimensions is where there is a 1:1 relationship between the base dim and the outrigger e.g. a Store Dim has a "First Open Date" which links to the Date Dim as an outrigger. This is not the situation you describe in your case - which seems to be closer to the "Multivalued Skill Keyword Attributes" issue Kimball describes in Chapter 9.

I'm not clear what the issue is with using a bridge table. Your Keyword Dimension would hold 100 records and the bridge table would hold the links between the customer and the Keyword Dim. I would also normally implement this as a Customer Factless Fact table so that the link from Customer to Bridge to Keyword is via the Fact and not between Dims.

The alternative approach outlined by Kimball is to put the Keywords in a delimited list in a single field. Probably the easiest to implement and then query using SQL but trickier to present sensibly to report writers through a BI tool such as OBIEE, SSRS, etc.

nick_white

Posts : 366
Join date : 2014-01-06
Location : London

View user profile

Back to top Go down

Re: Question about keyword outrigger

Post  ngalemmo on Thu Apr 24, 2014 2:08 pm

Common practice is to use a bridge table and a keyword dimension. The reason for this is, even if its not required now, the user may wish to add additional attributes to the keywords to provide semantic classifications to them (such as 'programming skill', 'network skill' and so on). Using the outrigger you describe does not provide an easy way to do that.
avatar
ngalemmo

Posts : 3000
Join date : 2009-05-15
Location : Los Angeles

View user profile http://aginity.com

Back to top Go down

Re: Question about keyword outrigger

Post  Sponsored content


Sponsored content


Back to top Go down

View previous topic View next topic Back to top

- Similar topics

 
Permissions in this forum:
You cannot reply to topics in this forum