Page 19 - Ai Robogenius

Some disadvantages of Primary data are as follows:

Often time-consuming and expensive.
Requires careful planning and design to avoid biases.
Secondary source of data

Secondary data is data that has already been collected, processed and published by other
individuals or organisations. Researchers use this data for new analysis or to supplement primary
data.
Examples of secondary data sources include:

Books and academic journals: Published research studies and scholarly articles.

Government reports: Census data, economic statistics, health records and official publications.
Websites and online databases: Public datasets, industry reports, open data portals like Kaggle
or WHO.

Newspapers and magazines: Articles and reports on current events and trends.

• UCI is a collection
• Countries like of databases,
Australia, EU, India, domain theories and
New Zealand and data generators in
Singapore are openly
sharing datasets on Dataset Search collaboration with
the University of
various portals
Massachusetts

• Kaggle is an online • This is a toolbox by
community of data .gov Google that can UCI
scientists where you search for data by
can access different datasets name Machine Learning Repository
types of data

Some advantages of secondary source data are as follows:

Easily accessible and often free or low-cost.

Saves time as data collection is already done.
Useful for historical or trend analysis.

Some disadvantages of secondary source data are as follows:

May not perfectly fit the current research question.
Possible issues with accuracy, reliability or outdatedness.

Lack of control over data quality and collection methods.
A list of government websites for data collection:

Stages of AI Project Cycle
17