Databases Reference
In-Depth Information
• Number of flights across time
• Number of transported passengers
• Amount of transported cargo (mail and freight)
• Most-used routes
At the same time, the application should allow the user to choose airline and aircraft
dimensions, as well as origin and destination airports, cities, and states.
Now that we have our goals clearly defined, let's move on to the construction phase.
Constructing the data model
The tutorial in this chapter is designed to focus mostly on creating the analysis
interface of our QlikView document. However, a fundamental part of QlikView
development is to construct an appropriate data model to support the various
analyses required in the application. The dataset we will work with in this chapter
will help us describe the most important concepts we need to consider when
building the data model.
What is a data model?
The heart of a QlikView application is its data model. It is composed of the different
source tables that contain the information and data used to measure a company's
performance. The data model is constructed by using QlikView's scripting language.
A correctly-built data model will associate all of its tables in a way which allows us
to manipulate the data however we like. This means that the creation of analysis
objects (charts) across different dimensions depends mainly on how the data model
is built and how its tables are associated (how they are linked to each other).
Loading the fact table
To start building our data model, we will load the fact table of our source data files
into QlikView.
A fact table is a table that contains the measurements across which we'll make the
analyses. The fact table is, at the same time, the central part of the data model.
A data model can contain more than one fact table. We'll deal
with the implications regarding schema design in
Chapter 4
,
Data Modeling
and
Chapter 8
,
Data Modeling Best Practices
.
