Build Relationships for Analytical Modeling  

Business Scenario

Welcome!


Building relationships in Microsoft Power BI is like a data analyst connecting information from different departments using common IDs.

These relationships help combine separate datasets into one meaningful business view.

Pre-Lab Preparation

Topic : Modelling Data in Power BI

1) Create relationships between tables

2) Define cardinality

3) Configure cross-filter direction

4) Manage relationships in model view  

Task 1: Create Relationshiop between tables

1

Without relationships, reports remain disconnected and insights become inaccurate. A data analyst uses relationships to create accurate, interactive, and reliable dashboards.

 

Download the below dataset and import into Power BI desktop  

Home Tab—> Get Data→Text/CSV option

git pull origin branchName

Git Pull

2

 Load this datasets into Power BI desktop,make sure all datasets are clean and already transformed  

3

Go to model view, available in left side pane of your screen

4

If relationship already exists, remove that and create new one by using

Home tab—> Manage Relationship—-> Select from table(column) check box —> delete —->Confirm Delete—>Close  

5

Zoom in relationship view, to see all tables, and bring them near to each other

Check the fact table and dimension table, arrange them in a circular manner. Keep Fact table in center of all dimension table

6

Go to Home tab—-> manage relationship —>New Relationship   

7

Select from table(fact table) and to table(dimension table), Select common column from table

Task 2: Define Cardinality

1

Once you select to table and from table with common column, you can select Cardinality type

2

Select Fact to dimension table is many to one (* : 1) type 

Task 3: Configure cross-filter direction

Select cross-filter direction either single or both, by type of relationship Typical Pattern for selection:  

Relationship Recommended
Dimension → FactSingle
Many-to-many Sometimes Both 
Bridge tables Often Both
Complex slicer interactionSometimes Both

So , We are selecting single cross filter direction—> Save Repeat the same steps for all tables in the model view.

Task 4: Manage Relationships in model view view

1

This is the final output screen for data modeling view for given dataset 

2

Before Close, if you want to edit or manage any relationship , you can select specific checkbox before that relationship and select edit option

3

Make the changes and then save changes

This is your final output

 

Great job!

You have successfully completed building data relation for analytical model lab

Checkpoint

Next-Lab Preparation

   Git Push

git push origin branchName

Topic : Prepare HR Workforce Dataset for Analysis

1) Import HR dataset

2) Clean missing values

3) Merge HR datasets

4) Append multiple datasets

Pb-5

By Content ITV

Pb-5

  • 5