Nerdy Nerdy Geeky Shit.

[jhameia]

I'm trying to explain the concept of Cognos 8 Business Intelligence tools, particularly the PowerCube (which is, in my document, a data cube).

To me, I would say "a data cube is a processor of intersecting search parameters used to calculate specific data elements in order to generate meaningful summary reports."

The Powercube site explains a datacube as a "multi-dimensional data set".

Now, a cube contains data within it, and a user can interact with the data in order to slice-and-dice down to something which has all the search paraameters s/he wants. For example, I could create a cube with the following dimensions:

Newspaper reports - which region I want - what newspapers I want - specific newspaper
intersecting with
Reports of economic losses - which region I want - which city I want
intersecting with
Reports of poverty - which region I want - which city I want
intersecting with
Reports of crime - which region I want - which city I want

And a couple other dimensions which I won't mention here... now, in a regular search engine, I could only do ONE of those searches at a time. And then within that I'd have to do another search for the second one, and so on.

Data cubes take these parameters and find all the specific reports I want for me to look at immediately.

And that's not it; that's just simplified in Jha's Concerns speak, for me to understand. Businesspeople don't normally search for social concern reports. They do the numbers thing. And data cubes enable them to take the numbers of more dimensions and process those numbers so that instead of getting a bunch of reports in which you have to figure out how to arrange them so they have any significance, they spit out reports which compare the numbers you want, all within the same report in a way which is pre-arranged for you to understand.

Let me see if I can find another example:

Statistics of poverty
X
statistics of crime
X
statistics of violence against women
X
statistics of domestic abuse
X
statistics of economic disparity
X
statistics of urba development

A data cube will comprise of all these statistics (that's a lot of numbers!) and I can mess around with specifics of EACH statistic (each of which has its own search parameters I can zero in onto) in order to come up with an analysis of the specific data I've taken a look at.

Hmmmmm.

This doesn't make sense to any of you guys, does it?

Comments

[salzara-tirwen.livejournal.com]

powercube, data set, right. Is there a reason they couldn't just call it a three (or n-) dimensional matrix?

[fantasyecho.livejournal.com]

It probably has to do with the marketing of it. "Multi-dimentional matrix" is a bit more of a mouthful than "powercube".

[salzara-tirwen.livejournal.com]

How about "search engine"? I mean, any good search engine can, say, look for things with X words in Y language.

(And your post makes perfect sense to me.)
If you're looking for a way to explain it to people, show them a table, and then tell them to imagine adding another layer?

[fantasyecho.livejournal.com]

Not really. This is going to take very specific data and drill-through all that data, with the possibility of zero-ing on a specific piece (say a specific report).

And ordinarily, I would use a Rubik's cube or a piece of cheese to demonstrate how it works, but that doesn't really translate well into text.

[fantasyecho.livejournal.com]

Also, it has to do with sets of data and trying to summarize its findings in order to contrast, compare and identify trends in the data.