Page 18 - Ai Robogenius

Examples:
Voice commands (e.g., “Hey Google”)

Music and soundtracks
Phone conversations
Audiobooks and podcasts

Characteristics of Good Data

To ensure effective use in AI modelling, data must be:
Relevant: Closely related to the problem being solved.

Accurate: Free from errors and inconsistencies.
Complete: All necessary fields are filled.

Timely: Up-to-date and reflects current conditions.
Balanced: No over-representation or under-representation of any group (to avoid bias).
Formatted: Structured correctly for easy analysis (e.g., tabular, JSON, XML, etc.)

Sources of Data

Data for any project, including AI projects, can be collected from two main sources: Primary and
Secondary sources. Understanding these sources helps ensure the data is reliable and suitable
for analysis.
Primary source of data

Primary data is original data collected directly from the source for the first time. This data is
specifically gathered to address the research problem or project requirements and has not been
previously published or used by others.

Primary source of data

Surveys Interviews Observations Experiments
Questionnaires One-on-one Systematic recording Controlled tests
distributed to a or group of behaviours or where variables
selected group of conversations events as they are manipulated to
people to gather their conducted to naturally occur, observe effects and
opinions, behaviours obtain in-depth without interference. outcomes.
or preferences. information.

Some advantages of Primary data are as follows:

Highly relevant and specific to the problem.

Control over the data collection process, ensuring accuracy and reliability.
Can provide up-to-date and real-time information.

AI RoboGenius - VIII
16