Artificial Intelligence

Class 11 Artificial Intelligence – Unit 5: Data Literacy – Data Collection to Data Analysis | Complete Notes

Class 11 · Artificial Intelligence

Unit 5: Data Literacy – Data Collection to Data Analysis

Class: XI | Subject: Artificial Intelligence | Unit: 5

Unit Overview

This unit introduces the importance of data literacy in Artificial Intelligence. Students learn how data is collected, classified, explored, analysed and visualized. The unit also introduces matrices, data preprocessing, modelling and evaluation, which form important foundations for AI and Machine Learning.

Complete Topics Covered in This Unit

What is Data Literacy?
Importance of Data in Artificial Intelligence
Data Collection – Primary and Secondary Sources
Levels of Measurement – Nominal, Ordinal, Interval and Ratio
Statistical Analysis of Data
Measures of Central Tendency – Mean, Median and Mode
Measures of Dispersion – Variance and Standard Deviation
Representation and Visualization of Data
Introduction to Matrices
Applications of Matrices in AI
Data Pre-processing
Data Cleaning, Transformation and Reduction
Data in Modelling and Evaluation
Classification and Regression Metrics

1. Learning Outcomes

By the end of this unit, students will be able to:

Explain the importance of data literacy in the context of AI.
Identify different data collection methods and understand their practical applications.
Understand mathematical concepts related to matrices, their operations and their role in AI.
Apply basic data analysis techniques to interpret data effectively.
Visualize data using different techniques to communicate meaningful insights.

2. What is Data Literacy?

Data can be defined as a representation of facts or instructions about an entity such as students, sports, animals or other objects that can be processed by humans or machines.

Data Literacy: Data literacy is the ability to find, understand, organize, analyse and use data effectively and ethically.

Why is Data Literacy Important in AI?

Artificial Intelligence is essentially data-driven. AI systems use large amounts of data to identify patterns, generate insights and make predictions.

Data literacy helps a person understand what the data means, identify useful information, analyse it correctly and make informed decisions.

Real-Life Analogy: Being data-literate is like having a map to navigate a big city. A map helps us understand where we are, where we need to go and which route to take. Similarly, data literacy helps us understand information and make smart decisions.

3. Data Collection

Data collection is the process of gathering information that can be used for analysis and decision-making.

In AI, data collection is particularly important because AI systems use data to identify recurring patterns and build predictive models.

A. Primary Sources

Primary data is data collected specifically for a particular analysis or purpose.

Method	Meaning	Example
Survey	Gathering information from a population using questionnaires or online forms.	Collecting consumer preferences for a new product.
Interview	Collecting information through direct communication with individuals.	Collecting employee feedback about job satisfaction.
Observation	Recording behaviour as it occurs naturally.	Observing children's play patterns to understand social dynamics.
Experiment	Manipulating variables to observe their effects.	Testing different advertising campaigns on different groups.

B. Secondary Sources

Secondary data refers to existing data that has already been collected and compiled for use.

Source	Meaning	Example
Web Scraping	Automatically collecting data from websites.	Collecting product prices from e-commerce websites.
Satellite Data Tracking	Monitoring information about the Earth's surface using satellites.	Tracking weather patterns or environmental changes.
Online Data Platforms	Websites that provide datasets for analysis.	Kaggle and GitHub.

4. Exploring Data – Levels of Measurement

Before analysing data, it is important to understand its values, type and characteristics. Data can be classified into four major levels of measurement.

Level	Description	Examples
Nominal	Data consists of categories with no order or ranking.	Eye colour, gender, car brands and cricket jersey numbers when used as names.
Ordinal	Categories have a meaningful order or rank, but the difference between values cannot be measured.	Hotel ratings, class rank and grades such as A, B and C.
Interval	Values are ordered and differences between values are meaningful, but there is no true zero.	Temperature in Celsius or Fahrenheit, shoe size and years.
Ratio	Similar to interval data but has a true zero, allowing meaningful ratios.	Weight, age and number of students.

Remember: The major difference between Interval and Ratio data is the presence of a true zero in ratio-level data.

5. Statistical Analysis of Data

Statistics helps transform raw observations into meaningful information that can be understood, interpreted and communicated.

A. Measures of Central Tendency

Measures of central tendency help identify the centre or typical value of a dataset.

Measure	Meaning	Example / Use
Mean	The mathematical average of the values.	Average marks of students in a test.
Median	The middle value when the data is arranged in order.	Finding the middle income value in an ordered dataset.
Mode	The value that occurs most frequently.	A bookstore can identify the most popular books using mode.

B. Measures of Dispersion

Measures of dispersion describe how widely the data values are spread around the centre.

Measure	Meaning
Variance	The average of the squared differences from the mean.
Standard Deviation	The square root of the variance.

Understanding Dispersion: A high variance indicates that the data points are widely spread out. A low standard deviation indicates that the data points are close to the mean.

Example: Suppose the heights of five dogs are recorded. Statistical measures can help determine the average height and how much individual dog heights differ from the group's average.

6. Representation and Visualization of Data

Data visualization presents information in a graphical or visual form. It makes large amounts of data easier to understand and helps identify trends, comparisons, distributions and relationships.

Visualization	Purpose	Example
Line Graph	Represents continuous data and shows trends over time.	Tracking a student's AI test marks over five tests.
Bar Graph	Compares discrete categories.	Comparing the number of books sold in different genres.
Histogram	Shows the distribution of data using intervals or bins.	Showing the distribution of student heights.
Scatter Graph	Shows the relationship or correlation between two variables.	Comparing study time with mathematics marks.
Pie Chart	Shows proportions or percentages as parts of a whole.	Showing monthly expenses as a percentage of income.

Quick Recall:
Line Graph → Trends over time
Bar Graph → Category comparison
Histogram → Data distribution
Scatter Graph → Relationship between variables
Pie Chart → Proportion or percentage

7. Introduction to Matrices

A matrix is a rectangular arrangement of numbers organised into rows and columns.

Matrices are extremely important in Artificial Intelligence because large amounts of numerical information can be represented and processed in matrix form.

Applications of Matrices in AI

AI Application	Role of Matrices
Digital Images	Images are stored as matrices containing pixel values, generally represented by values from 0 to 255.
Recommender Systems	Matrices can represent relationships between users and the products they purchase.
Natural Language Processing	Vectors, which can be considered one-dimensional matrices, are used to represent the distribution of words in documents.

Matrix Representation

A matrix can be represented using rows and columns.

[ 10   20   30 ]
[ 40   50   60 ]

The matrix above contains 2 rows and 3 columns. Therefore, its order is 2 × 3.

8. Data Pre-processing

Raw data is often incomplete, inconsistent or difficult to use directly. Therefore, data must be prepared before it is supplied to an AI model. This process is called data preprocessing.

Data Pre-processing: The process of preparing raw data so that it becomes suitable for analysis and AI modelling.

Major Steps of Data Pre-processing

Step	Meaning	Purpose
1. Data Cleaning	Handling missing values and noisy or incorrect data.	Improves the quality and reliability of data.
2. Data Transformation	Changing data into a suitable format or scale.	Helps ensure that different features are on a similar scale.
3. Data Reduction	Reducing the volume of data while preserving its important information.	Makes processing more efficient.

Real-Life Analogy: Data preprocessing is like sorting a giant bag of mixed candies. Before finding specific flavours, we first need to clean, separate and organize the candies. Similarly, AI needs clean and organized data before analysis.

9. Data in Modelling and Evaluation

Modelling

In AI, datasets are used to train models. The model learns patterns from the available data and uses these patterns to make predictions or decisions.

Evaluation

After a model has been trained, it must be tested to determine how well it performs. This process is called model evaluation.

Problem Type	Evaluation Metrics
Classification	Accuracy, Precision and Recall
Regression	Mean Squared Error (MSE) and R-squared

10. Classification Metrics

Accuracy

Accuracy measures how many predictions made by a classification model are correct out of the total predictions.

Accuracy = Correct Predictions / Total Predictions

Precision

Precision indicates how many of the instances predicted as positive were actually positive.

Recall

Recall indicates how many of the actual positive instances were correctly identified by the model.

11. Regression Metrics

Mean Squared Error (MSE)

Mean Squared Error measures the average of the squared differences between actual values and predicted values.

MSE = Average of Squared Prediction Errors

A lower MSE generally indicates that the predictions are closer to the actual values.

R-squared

R-squared is a measure used in regression to understand how well the model explains the variation in the data.

12. Data Literacy in the AI Workflow

Data literacy plays an important role throughout the AI development process.

Stage	Role of Data
Data Collection	Gather relevant information from primary or secondary sources.
Data Exploration	Understand the nature, values and measurement levels of the data.
Data Analysis	Use statistical techniques to identify patterns and insights.
Data Visualization	Represent information graphically for easier interpretation.
Data Pre-processing	Clean, transform and reduce data before modelling.
Modelling	Use datasets to train an AI model.
Evaluation	Measure the performance of the trained model.

13. Quick Revision Table

Term	Key Idea
Data	Representation of facts or instructions that can be processed.
Data Literacy	Ability to find, understand, organize, analyse and use data effectively and ethically.
Primary Data	Data collected specifically for a particular purpose.
Secondary Data	Existing data collected and compiled earlier.
Nominal	Categories without order.
Ordinal	Ordered categories.
Interval	Ordered data without a true zero.
Ratio	Ordered data with a true zero.
Mean	Mathematical average.
Median	Middle value of ordered data.
Mode	Most frequently occurring value.
Variance	Average of squared differences from the mean.
Standard Deviation	Square root of variance.
Matrix	Rectangular arrangement of numbers in rows and columns.
Data Cleaning	Handling missing and noisy data.
Data Transformation	Changing data into a suitable format or scale.
Data Reduction	Reducing data volume while retaining important information.
Accuracy	Measures correct classification predictions.
Precision	Measures correctness among predicted positive cases.
Recall	Measures correctly identified actual positive cases.
MSE	Measures average squared prediction error in regression.
R-squared	Measures how well a regression model explains variation in data.

14. Exam-Oriented Key Points

Data literacy is essential because AI systems are data-driven.
Primary data is collected specifically for analysis, while secondary data already exists.
Nominal data has categories without order.
Ordinal data has ordered categories.
Interval data has meaningful differences but no true zero.
Ratio data has a true zero.
Mean, Median and Mode are measures of central tendency.
Variance and Standard Deviation are measures of dispersion.
Line graphs are useful for trends over time.
Bar graphs are useful for comparing categories.
Histograms represent data distributions.
Scatter graphs show relationships between two variables.
Pie charts represent proportions or percentages.
A matrix is an arrangement of numbers in rows and columns.
Digital images can be represented as matrices of pixel values.
Data preprocessing includes cleaning, transformation and reduction.
Classification can be evaluated using accuracy, precision and recall.
Regression can be evaluated using MSE and R-squared.

15. One-Line Revision

Data → Collection → Exploration → Analysis → Visualization → Pre-processing → Modelling → Evaluation

This sequence represents the major journey of data from collection to its use in an Artificial Intelligence system.