Page 112 - FULL REPORT 30012024

ii. Transform

The transform phase in the ETL process is a critical juncture where the

dataset undergoes a series of queries to ensure its structure and content
are primed for effective analysis in Power BI. This stage involves a

meticulous examination and refinement of the dataset, pivotal to

ascertaining the data's accuracy, consistency, and relevance. The first
steps in this phase are data validation procedures, which involve

running targeted queries that separate records based on certain
attributes, like where they are located geographically and when they

were created. Such queries substantiate the data's categorization, which

is crucial for subsequent aggregation processes in Power BI, as
delineated in Figure 4.31, showcasing the data filtering by country

'Malaysia'.

Figure 4.31 Query for data filtering by country.

Moreover, the transform step comprises an integrity check to discover

any data abnormalities, confirming the dataset's quality. In order to
recognise future health emergencies and locations requiring rapid care,

queries that retrieve data based on predetermined circumstances, such
as death rates above particular thresholds, are an example of this

95