Data Acquisition | Artificial Intelligence for Class 10 PDF Download

Understanding data acquisition

Data Acquisition consists of two words:

• Data: Refers to raw facts, figures, or pieces of information collected for reference or analysis.
• Acquisition: Involves obtaining data for a specific project.

The process of gathering data from relevant sources is termed data acquisition.

Classification of Data

Now, observe the following diagram for data classification. We will discuss each category in detail:

Basic Data

• Numeric Data: Primarily used for calculations. Numeric data can be divided into the following:
  • Discrete Data: Consists only of integer numeric values without any decimal or fractional parts. Countable data falls into this category, such as 132 customers or 126 students.
  • Continuous Data: Includes data within any range, encompassing uncountable values. Examples include 10.5 KGS or 100.50 Kms.
• Text Data: Mainly used to represent names, collections of words, phrases, and other textual information.

Structural Classification

The data used to train the model or already present in the system may follow specific constraints, rules, or unique patterns, and is considered structural data.
The structure classification is divided into three categories:

• Structured Data: This data follows a specific pattern or set of rules, has a simple structure, and is stored in specific formats such as tabular form. Examples include a cricket scoreboard, a school timetable, or an exam datasheet.
• Unstructured Data: This type of data does not follow any specific pattern or constraints and can be stored in various forms. Most of the data in the world is unstructured. Examples include YouTube videos, Facebook photos, or dashboard data from any reporting tool.
• Semi-Structured Data: This is a blend of both structured and unstructured data. Some parts of the data may have a defined structure, like a database, while other parts may include markers and tags to identify the data's structure.

Other Classification

This classification is further divided into the following branches:

• Time-Stamped Data: This structure aids the system in predicting the next best action by following a specific time order to define the sequence. The time can refer to when the data was captured, processed, or collected.
• Machine Data: This is the result or output of a specific program, system, or technology. It includes data related to a user's interaction with the system, such as logged-in session data, specific search records, and user engagement like comments, likes, and shares.
• Spatiotemporal Data: This data includes information related to geographical location and time. It records the location through GPS and includes time-stamped data where the event is captured or data is collected.
• Open Data: Data that is freely available for everyone to use and reuse.
• Real-time Data: Data that is available as the event occurs.
• Big Data: This term refers to data that cannot be stored by any traditional system or data collection software like DBMS or RDBMS. Big data is a broad and complex topic.

Data Features

Data features refer to the types of data you want to collect.
Two key terms are associated with this:

• Training Data: The data collected through the system is known as training data. Essentially, it is the input provided by the user to the system.
• Testing Data: The processed data or result data set is known as testing data. Essentially, it is the output of the data.