Page 14 - CT_AI_Class-7

Based on Data Structure

Data structure refers to how data is organised, such as in tables, files or unstructured formats.
Datasets can be classified based on their organisation and structure into the following types:

 Structured data: Structured data refers to data that is neatly
organised in a fixed format, such as rows and columns in tables
or databases, making it easy for machines to store, process and
analyse. An example of structured data is spreadsheets or databases
in which data is stored such as student records with names, marks
and roll numbers.

 Semi-structured data: Semi-structured data is a type of data that does
not follow a strict table format but still contains some organisational
elements, such as tags or labels, which help machines understand and
process it. An example of semi-structured data is an email which contain
structured elements such as the sender, recipient, timestamp and subject

line, which follow a predictable format. However, the body of the email
itself is unstructured, as it can contain free-form text, images and
attachments.

 Unstructured data: Unstructured data refers to data that does not
have any predefined format or organisation, such as text, images,
videos or audio, making it more complex for machines to analyse.

An example of unstructured data is a collection of photos and videos
shared on social media does not follow a fixed structure.

 Hybrid data: Hybrid datasets are a combination of structured, semi-structured and
unstructured data. They contain multiple types of data within a single dataset.

This type of dataset is commonly found in real-world applications, where different forms of
data are collected and used together for better insights and decision-making. An example of
hybrid datasets is e-commerce platforms combining transaction data, product descriptions and
customer reviews.

12 Artificial Intelligence (CT & AI)-VII