Artificial Intelligence (AI)

Artificial Intelligence (AI) is a technology that enables computers and machines to perform tasks that normally require human intelligence, such as understanding language, recognizing images, and solving problems.

Examples of AI [Mention any 1 or 2 Examples during writing exam]

• Google Assistant: Understands voice commands and responds to users.
• ChatGPT: Answers questions, explains topics, and helps users write content.
• Self-Driving Cars: Use AI to recognize roads, traffic signs, and other vehicles.
• YouTube: Recommends videos based on users' interests.
• Face Recognition: Identifies people by analyzing their facial features.
• Online Chatbots: Answer customers' questions on websites.
• Medical Diagnosis: Helps doctors identify diseases by analyzing medical images and data.
• Robots: Use AI to perform tasks such as sorting objects and working in factories.

Machine Learning (ML)

Machine Learning (ML) is a branch of Artificial Intelligence that enables computers to learn from data, identify patterns, and make predictions or decisions without being programmed for every task.

Examples of ML [Mention any 1 or 2 Examples during writing exam]

• YouTube: Recommends videos based on users' interests and viewing history.
• Email Spam Filters: Learn to identify and separate spam emails from normal emails.
• Online Shopping: Recommends products based on users' previous searches and purchases.
• Netflix: Recommends movies and shows based on users' viewing habits.

Difference Between Artificial Intelligence (AI) and Machine Learning (ML)

Artificial Intelligence (AI)	Machine Learning (ML)
AI is a technology that enables machines to perform tasks that normally require human intelligence.	ML is a branch of AI that enables machines to learn from data.
It focuses on making machines perform intelligent tasks.	It focuses on learning patterns from data and making predictions.
AI can use different methods, including rules, logic, and machine learning.	ML uses data and algorithms to learn patterns.
AI is the broader field.	ML is a part of AI.
Examples: Google Assistant, ChatGPT, Self-Driving Cars, Face Recognition, Rule-Based Chatbots.	Examples: YouTube Recommendations, Email Spam Filters, Netflix Recommendations, Product Recommendations, Fraud Detection.

Types of Machine Learning

Machine Learning (ML) is a branch of Artificial Intelligence that enables computers to learn from data and make predictions or decisions. It has three main types:

• Supervised Learning
• Unsupervised Learning
• Reinforcement Learning: The computer learns through trial and error by receiving rewards for good actions and penalties for incorrect actions.

Difference Between Supervised and Unsupervised Learning

Supervised and unsupervised learning are two important types of machine learning. The main difference is that supervised learning uses labeled data, while unsupervised learning uses unlabeled data.

Supervised Learning	Unsupervised Learning
It learns from labeled data, where the correct answers are already known.	It learns from unlabeled data, where the correct answers are not provided.
The computer learns by comparing its predictions with the correct answers.	The computer identifies patterns, similarities, and groups in the data on its own.
It is mainly used to predict results or classify information into known categories.	It is mainly used to discover hidden patterns and group similar information.
Example: Teaching a computer using labeled pictures of cats and dogs so it can identify them in new pictures.	Example: Giving a computer pictures of different animals without labels so it can group similar animals together.

Practical Applications of AI/ML

1. Smart Assistants

Smart assistants use AI and ML to understand voice commands, answer questions, and perform different tasks. They help users set alarms, search for information, make calls, send messages, and control smart devices.

Examples: Siri, Google Assistant, and Alexa.

2. Recommendation Systems

Recommendation systems use AI and ML to suggest videos, products, music, movies, and other content based on users' interests, searches, and previous activities. They help users find content they may like.

Examples: YouTube recommends videos, Netflix suggests movies, and Daraz recommends products based on users' interests and activities.

Bias, Privacy, and Responsibility in AI

1. Bias in AI

Bias in AI occurs when an AI system produces unfair results because of the data it learns from.

Example: If an AI system is trained only on English or male voices, it may not work well with Urdu speakers or female voices.

2. Privacy and Data Use

AI systems often use personal information, such as names, locations, and user preferences. This information must be protected from misuse.

Example: An app that tracks a user's location without permission can misuse their personal information.

Best Practices

• Ask users for permission before collecting their data.
• Protect data using encryption.
• Collect only the data that is needed.

3. Responsibility of Humans

Humans are responsible for developing, using, and monitoring AI systems. They must check AI decisions and correct mistakes to prevent harm.

Example: If a facial recognition system identifies the wrong person, developers must investigate the error and improve the system to prevent similar mistakes.

Supervised and Unsupervised Learning Algorithms

1. Linear Regression (Supervised Learning)

Linear Regression is a supervised learning algorithm that predicts future values by finding a straight-line relationship between existing data. It studies how one value changes with another to make predictions.

Example: Predicting a student's exam score based on the number of hours they study. Generally, students who study more hours may achieve higher scores.

2. Decision Trees (Supervised Learning)

Decision Trees are supervised learning algorithms that make predictions or decisions by following a tree-like structure of questions and conditions.

Example: Predicting whether a student will pass or fail based on their exam preparation and previous marks.

3. K-Means Clustering (Unsupervised Learning)

K-Means Clustering is an unsupervised learning algorithm that groups similar data into different clusters based on their features, without using predefined labels.

Example: Grouping customers based on their shopping habits, such as customers who buy frequently, occasionally, or rarely.

Evaluating Model Performance

Model Performance Evaluation: It is the process of checking how accurately a machine learning model makes predictions.

Confusion Matrix

A confusion matrix is a table used to compare a model's predictions with the actual results. It helps check the performance and quality of a machine learning model by showing its correct and incorrect predictions.

A confusion matrix contains four values:

• True Positive (TP): The model correctly predicts a positive result.
• True Negative (TN): The model correctly predicts a negative result.
• False Positive (FP): The model predicts a positive result, but the actual result is negative.
• False Negative (FN): The model predicts a negative result, but the actual result is positive.

Example of a Confusion Matrix

Suppose an AI model predicts whether students will pass or fail an exam. After comparing the predictions with the actual results of 100 students, the following confusion matrix is obtained:

Model Predictions	Actual Results
	Actual Pass	Actual Fail
Predicted Pass	TP = 40	FP = 10
Predicted Fail	FN = 5	TN = 45

• TP = 40: The model correctly predicts that 40 students will pass.
• TN = 45: The model correctly predicts that 45 students will fail.
• FP = 10: The model predicts that 10 students will pass, but they actually fail.
• FN = 5: The model predicts that 5 students will fail, but they actually pass.

Accuracy

Accuracy measures the percentage of all predictions that are correct. It is calculated by dividing the total number of correct predictions by the total number of predictions and multiplying by 100.

Formula:

Accuracy = (TP + TN) / (TP + TN + FP + FN) × 100

Calculation:

Accuracy = (40 + 45) / (40 + 45 + 10 + 5) × 100

Accuracy = 85 / 100 × 100

Accuracy = 85%

Result: The model has an accuracy of 85%, meaning it correctly predicts the results for 85 out of 100 students.

Exercise: Short Answer Questions

1. What is the difference between AI and ML?

Already mentioned in the notes.

2. Give one real-life example of AI used in Pakistan.

One real-life example of AI used in Pakistan is a chatbot on a bank's website that answers customers' questions automatically.

3. How does supervised learning work?

Supervised learning works by training a model using labeled data, where the correct answers are already known. The model learns patterns from this data and uses them to make predictions on new data.

4. Define confusion matrix.

A confusion matrix is a table used to compare a machine learning model's predictions with actual results. It helps evaluate the model's performance by showing correct and incorrect predictions through True Positive (TP), True Negative (TN), False Positive (FP), and False Negative (FN).

5. How can businesses use AI?

Businesses can use AI to answer customer questions through chatbots, recommend products, analyze customer data, detect fraud, and improve decision-making.

6. Give one example each of a smart assistant and a recommendation system.

• Smart Assistant: Google Assistant helps users answer questions and perform tasks through voice commands.
• Recommendation System: YouTube recommends videos based on users' interests and viewing history.

7. What kind of data is used in unsupervised learning?

Unsupervised learning uses unlabeled data, which has no predefined labels or correct answers. The computer finds similarities and groups similar data on its own.

Example: Giving a computer pictures of different animals without names. The computer groups similar animals, such as cats and dogs, based on their features.

Exercise: Long Answer Questions

1. Explain supervised and unsupervised learning with suitable examples.

Already mentioned in the notes.

2. Explain how model performance is evaluated in machine learning using the following metrics:

a) Confusion Matrix

A confusion matrix is a table used to compare a model's predicted results with the actual results. It helps evaluate the performance and quality of a machine learning model.

It contains four values:

• True Positive (TP): The model correctly predicts a positive result.
• True Negative (TN): The model correctly predicts a negative result.
• False Positive (FP): The model predicts a positive result, but the actual result is negative.
• False Negative (FN): The model predicts a negative result, but the actual result is positive.

b) Accuracy

Accuracy measures the percentage of all predictions that are correct. It is calculated by dividing the number of correct predictions by the total number of predictions and multiplying by 100.

Formula:

Accuracy = (TP + TN) / (TP + TN + FP + FN) × 100

c) Precision

Precision measures how many of the positive predictions made by a model are actually positive. It helps determine how reliable the model's positive predictions are.

Formula:

Precision = TP / (TP + FP) × 100

3. Numerical Problem: Confusion Matrix and Accuracy

A model predicts whether students will pass (1) or fail (0). Out of 20 students, the results are:

• True Positives (TP) = 8
• True Negatives (TN) = 6
• False Positives (FP) = 4
• False Negatives (FN) = 2

(a) Fill the Confusion Matrix

The confusion matrix is completed by placing the given values according to the predicted and actual results.

Predicted Results	Actual Results
	Pass (1)	Fail (0)
Pass (1)	TP = 8	FP = 4
Fail (0)	FN = 2	TN = 6

(b) Calculate Accuracy

Formula:

Accuracy = (TP + TN) / (TP + TN + FP + FN) × 100

Step 1: Substitute the values.

Accuracy = (8 + 6) / (8 + 6 + 4 + 2) × 100

Step 2: Add the values.

Accuracy = 14 / 20 × 100

Step 3: Calculate the percentage.

Accuracy = 0.7 × 100

Final Answer: Accuracy = 70%

Result: The model correctly predicts the results for 14 out of 20 students, giving an accuracy of 70%.

Class 10 Computer Science Chapter 6 MCQs: Introduction to AI and Machine Learning

Test Yourself: Interactive MCQs (AI and Machine Learning)

Class 10 Computer Science Chapter 6 MCQs: Introduction to AI and Machine Learning

1. What is Artificial Intelligence (AI)?

a. Technology that enables machines to perform tasks requiring human intelligence b. The process of creating computer hardware c. A method of designing websites only d. A language used only for writing programs

2. What is Machine Learning (ML)?

a. A method of repairing computers b. A branch of AI that learns patterns from data c. A process of designing computer networks d. A type of computer hardware

3. Which of the following is an example of AI?

a. A simple wooden table b. A manual pencil sharpener c. A voice assistant that understands commands d. A regular notebook

4. Which of the following is a smart assistant?

a. Microsoft Paint b. Notepad c. Calculator without AI features d. Google Assistant

5. What is a common use of smart assistants?

a. Understanding voice commands and answering questions b. Printing paper without a printer c. Replacing all computer hardware d. Increasing the physical size of a computer

6. Which technology helps YouTube recommend videos?

a. Word processing b. AI and Machine Learning c. Computer hardware repair d. Manual data entry only

7. What is the purpose of a recommendation system?

a. To delete all user information b. To repair damaged computer parts c. To suggest items based on user interests and activities d. To create computer hardware

8. Which platform uses recommendation systems?

a. A basic wall clock b. A paper notebook c. A manual typewriter d. Netflix

9. What is a chatbot?

a. A computer program that communicates with users b. A device used only to print documents c. A computer storage cable d. A type of computer monitor

10. What is image recognition?

a. Creating a keyboard layout b. Identifying objects, faces, or patterns in images c. Printing text on paper d. Connecting a computer to electricity

11. What does AI bias mean?

a. The computer has no storage space b. The computer cannot connect to the internet c. An AI system produces unfair results for some people or groups d. A computer has a broken keyboard

12. What can cause bias in an AI system?

a. Using a clean computer screen b. Installing a new keyboard c. Using a larger monitor d. Training the system on incomplete or unbalanced data

13. What does data privacy mean?

a. Protecting personal information from unauthorized access or use b. Sharing everyone's personal information publicly c. Deleting every computer program d. Making all files available to everyone

14. Which practice helps protect personal data?

a. Sharing passwords with strangers b. Using encryption to protect sensitive information c. Publishing private information online d. Collecting unnecessary personal information

15. Who should help ensure an AI system is designed responsibly?

a. Only the computer monitor manufacturer b. Only the internet service provider c. Developers and the people responsible for the system d. No one

16. What is Machine Learning mainly used for?

a. Increasing the weight of a computer b. Painting computer cases c. Replacing electricity d. Learning patterns from data to make predictions or decisions

17. What is supervised learning?

a. Learning from labeled data with known answers b. Learning without using any data c. Grouping data without predefined labels d. A method of repairing hardware

18. What is unsupervised learning?

a. Learning only from labeled examples b. Finding patterns in data without predefined labels c. Learning only from a teacher's spoken instructions d. A method of creating websites

19. What is labeled data?

a. Data that has been permanently deleted b. Data that contains no information c. Data that includes known answers or categories d. Data stored only on paper

20. Which type of learning uses unlabeled data?

a. Supervised learning b. Manual learning only c. Traditional data entry d. Unsupervised learning

21. Which is an example of supervised learning?

a. Predicting student marks using study hours and known scores b. Grouping customers without predefined categories c. Arranging files randomly d. Turning off a computer

22. Which is an example of unsupervised learning?

a. Predicting a house price using known prices b. Grouping customers according to similar shopping habits c. Predicting exam marks from labeled examples d. Checking a password manually

23. What is a machine learning algorithm?

a. A physical part of a computer b. A type of computer screen c. A set of steps or rules used to learn patterns from data d. A device used to print photographs

24. What does Linear Regression predict?

a. Only image colors b. Computer hardware failures only c. Categories without using data d. Numerical values based on relationships between variables

25. Which is an example of Linear Regression?

a. Predicting student scores based on study hours b. Grouping customers into clusters c. Identifying a face in a photograph only d. Sorting computer cables by color

26. What is a Decision Tree?

a. A device used to store files b. A model that makes decisions using a tree-like structure of questions or conditions c. A method of encrypting every file d. A tool used only for drawing pictures

27. Which problem can a Decision Tree help solve?

a. Increasing the screen size b. Repairing a broken mouse physically c. Predicting whether a student will pass or fail d. Charging a laptop battery

28. What is clustering?

a. Deleting data from a computer b. Converting text into printed pages c. Giving every data point a different random value d. Grouping similar data points together

29. K-Means is an example of which type of algorithm?

a. Unsupervised learning algorithm b. Supervised learning algorithm only c. Word processing algorithm d. Computer hardware algorithm

30. What is the purpose of K-Means Clustering?

a. To predict values using a straight line only b. To divide data into groups based on similarity c. To encrypt passwords d. To design computer networks

31. What is model performance evaluation?

a. The process of changing a computer's appearance b. The process of collecting computer parts c. The process of checking how accurately a model makes predictions d. The process of installing an operating system

32. What is a confusion matrix?

a. A table of computer prices b. A list of programming languages c. A diagram of computer hardware d. A table comparing actual results with predicted results

33. What does True Positive (TP) mean?

a. The model correctly predicts a positive result b. The model incorrectly predicts a positive result c. The model correctly predicts a negative result d. The model incorrectly predicts a negative result

34. What does True Negative (TN) mean?

a. The model incorrectly predicts a positive result b. The model correctly predicts a negative result c. The model correctly predicts a positive result d. The model misses a positive result

35. What does False Positive (FP) mean?

a. The model correctly predicts a positive result b. The model correctly predicts a negative result c. The model predicts positive when the actual result is negative d. The model predicts negative when the actual result is positive

36. What does False Negative (FN) mean?

a. The model correctly predicts a negative result b. The model correctly predicts a positive result c. The model predicts positive when the actual result is negative d. The model predicts negative when the actual result is positive

37. Which formula is used to calculate accuracy?

a. (TP + TN) / (TP + TN + FP + FN) × 100 b. (FP + FN) / (TP + TN) × 100 c. (TP + FP) / (TN + FN) × 100 d. (TP + TN + FP + FN) × 100

38. A model makes 85 correct predictions out of 100. What is its accuracy?

a. 50% b. 85% c. 90% d. 100%

39. If TP = 40, TN = 45, FP = 10, and FN = 5, how many predictions are correct?

a. 10 b. 45 c. 85 d. 100

40. Using TP = 40, TN = 45, FP = 10, and FN = 5, what is the model's accuracy?

a. 40% b. 45% c. 80% d. 85%

FAQs

Class 10 Computer Science Chapter 6: AI and Machine Learning FAQs

1. What is Artificial Intelligence (AI)?

Artificial Intelligence (AI) is a technology that enables computers and machines to perform tasks that normally require human intelligence, such as understanding language, recognizing images, and solving problems.

2. What is Machine Learning (ML)?

Machine Learning (ML) is a branch of AI that allows computers to learn from data, identify patterns, and make predictions or decisions.

3. What is the difference between AI and ML?

AI enables machines to perform tasks that require human intelligence, while ML is a branch of AI that helps machines learn from data and improve their predictions.

4. What are some common applications of AI?

Common applications of AI include smart assistants, chatbots, image recognition, recommendation systems, and voice recognition.

5. What are smart assistants?

Smart assistants are AI-powered tools that understand voice commands and help users perform tasks. Examples include Siri, Google Assistant, and Alexa.

6. What is a recommendation system?

A recommendation system suggests videos, products, music, or other content based on users' interests and activities. Examples include YouTube, Netflix, and Daraz.

7. What is AI bias?

AI bias occurs when an AI system produces unfair results for certain people or groups, often because of incomplete or unbalanced training data.

8. Why is data privacy important in AI?

Data privacy is important because AI systems may use personal information. Protecting this information helps prevent unauthorized access and misuse.

9. What is supervised learning?

Supervised learning is a type of machine learning that uses labeled data, where the correct answers are already known. For example, predicting student marks using study hours and previous scores.

10. What is unsupervised learning?

Unsupervised learning is a type of machine learning that identifies patterns and groups similar data without predefined labels. For example, grouping customers by their shopping habits.

11. What is Linear Regression?

Linear Regression is a machine learning algorithm used to predict numerical values based on the relationship between variables. For example, predicting student marks from study hours.

12. What is a Decision Tree?

A Decision Tree is a machine learning model that makes predictions by following a tree-like structure of questions and conditions. For example, it can predict whether a student will pass or fail.

13. What is K-Means Clustering?

K-Means Clustering is an unsupervised learning algorithm that divides data into groups based on similarities between data points.

14. What is a confusion matrix?

A confusion matrix is a table that compares actual results with predicted results. It helps evaluate a machine learning model by showing correct and incorrect predictions using TP, TN, FP, and FN.

15. What is model accuracy, and how is it calculated?

Model accuracy measures the percentage of predictions a model gets correct. It is calculated using the formula: Accuracy = (TP + TN) / (TP + TN + FP + FN) × 100, where TP is True Positive, TN is True Negative, FP is False Positive, and FN is False Negative.

Download Notes

Download PDF