Table of Contents
UNIT–1: Introduction to AI Project Cycle & AI Ethics
CHAPTER 1- Artificial Intelligence: Basics and Project Cycle
1️⃣ ARTIFICIAL INTELLIGENCE (AI) – BASIC UNDERSTANDING
🔹 What is Artificial Intelligence?
Artificial Intelligence (AI) is a branch of computer science that enables machines and computers to think, learn, and make decisions like humans.
In simple words: “AI is the ability of a machine to behave intelligently.“
An AI system can:
- Learn from data
- Identify patterns
- Make decisions
- Improve its performance over time
💡 Important Point:
AI does not mean robots only. Many AI systems work silently inside computers, mobiles, and apps.
🔹 AI as a Problem-Solving Technology
AI is mainly used to solve real-world problems that are:
- Too complex for humans
- Require analysis of large data
- Need quick and accurate decisions
AI helps by:
- Studying the problem
- Collecting and analyzing data
- Finding the best possible solution
📌 Example:
Humans cannot analyze millions of online searches in seconds, but AI can.
🔹 Examples of AI in Daily Life
AI is already part of our everyday life:
- Search Engines
- Google suggests results based on our searches
- Auto-complete suggestions
- Voice Assistants
- Google Assistant, Alexa, Siri
- Understand voice commands and respond
- Recommendation Systems
- YouTube video suggestions
- Netflix movie recommendations
- Online shopping product suggestions
- Smartphones
- Face unlock
- Camera filters
- Spam call detection
📌 These systems learn from user data and behavior, which makes them AI-based.
2️⃣ AI PROJECT CYCLE – OVERVIEW
🔹 What is an AI Project Cycle?
An AI Project Cycle is a step-by-step process used to build an AI solution for a real-world problem.
Just like:
- A science experiment has steps
- A cooking recipe follows steps
👉 AI projects also follow fixed steps to ensure accuracy and success.
🔹 Why Do AI Projects Follow a Step-by-Step Cycle?
AI projects follow a cycle because:
- It helps understand the problem clearly
- It avoids confusion and mistakes
- It ensures proper use of data
- It improves the final result
📌 Skipping any step may lead to:
- Wrong results
- Biased AI
- Failed AI solutions
🔹 Relationship Between Real-World Problems and AI Solutions
AI projects always start with a real-life problem, such as:
- Crop failure
- Disease detection
- Pollution monitoring
- Traffic management
AI uses:
- Data
- Technology
- Logic
to create a solution for the problem.
👉 No problem = No AI project
3️⃣ STAGES OF AI PROJECT CYCLE (INTRODUCTORY LEVEL)
The AI Project Cycle has five main stages:
🔸 Stage 1: Problem Scoping
This is the most important stage.
Here we:
- Identify the problem
- Understand who is affected
- Decide what needs to be solved
📌 If the problem is not clear, the AI solution will fail.
🔸 Stage 2: Data Acquisition
Data is the fuel of AI.
In this stage:
- Data is collected from different sources
- The data may be:
- Text
- Numbers
- Images
- Audio
📌 Without data, AI cannot learn.
🔸 Stage 3: Data Exploration
After collecting data, we:
- Study the data
- Find patterns
- Remove errors or irrelevant data
This helps in understanding:
- What the data tells us
- What information is useful
🔸 Stage 4: Modelling
In this stage:
- AI models are created
- The machine is trained using data
- The AI learns patterns
📌 This is where actual intelligence is developed.
🔸 Stage 5: Evaluation
In the final stage:
- The AI solution is tested
- Accuracy is checked
- Performance is evaluated
If results are not good:
- The process is repeated
- Improvements are made
4️⃣ 4Ws FRAMEWORK FOR PROBLEM SCOPING
To clearly define a problem, we use the 4Ws Framework.
🔹 Who is Affected?
- Who faces the problem?
- Farmers, students, doctors, common people?
📌 Identifying affected people helps focus the solution.
🔹 What is the Problem?
- What exactly is going wrong?
- What needs to be solved?
📌 The problem should be clear and specific.
🔹 Where Does It Occur?
- In a village, city, school, hospital, or country?
- At a local or global level?
📌 Location helps decide the scale of the solution.
🔹 Why Does It Matter?
- Why should this problem be solved?
- What happens if it is ignored?
📌 This shows the importance of the problem.
5️⃣ AI FOR SUSTAINABLE DEVELOPMENT GOALS (SDGs
🔹 What Are SDGs?
Sustainable Development Goals (SDGs) are global goals set by the United Nations to:
- Improve quality of life
- Protect the environment
- Ensure equality and development
There are 17 SDGs, such as:
- No Poverty
- Zero Hunger
- Good Health
- Quality Education
- Climate Action
🔹 Understanding SDGs in Simple Terms
SDGs aim to:
- Solve major world problems
- Create a better future for everyone
- Balance development and environment
🔹 Role of AI in Solving Global Problems
AI helps in achieving SDGs by:
- Predicting crop diseases
- Monitoring climate change
- Improving healthcare diagnosis
- Managing resources efficiently
📌 AI supports smart and sustainable solutions.
6️⃣ SUGGESTED ACTIVITIES (EXPLAINED)
🔹 SDG-Based AI Problem Identification
- Students identify a real problem related to an SDG
- Apply AI Project Cycle steps
🔹 Case Study 1: Coffee Production System
- Problem: Low coffee yield
- AI use: Predict soil health and weather
- SDG: Zero Hunger, Decent Work
🔹 Case Study 2: Earth-Like Exoplanet Discovery
- Problem: Finding habitable planets
- AI use: Analyze space data
- SDG: Innovation and Scientific Progress
🔹 Group Discussion Using 4Ws Framework
- Students work in groups
- Apply Who, What, Where, Why
- Present AI-based solutions
CHAPTER 2- Data and Problem Scoping in Artificial Intelligence
1️⃣ DETAILED UNDERSTANDING OF EACH AI PROJECT STAGE
In CHAPTER 1, we learned the names of the stages of the AI Project Cycle.
In CHAPTER 2, we understand why each stage is important and what happens if it is ignored.
The five stages are:
- Problem Scoping
- Data Acquisition
- Data Exploration
- Modelling
- Evaluation
Each stage plays a critical role in building a successful AI system.
2️⃣ WHY PROBLEM SCOPING IS IMPORTANT
🔹 Meaning of Problem Scoping
Problem Scoping means:
- Clearly defining the problem
- Understanding the needs of the people affected
- Setting clear goals for the AI solution
🔹 Importance of Problem Scoping
Problem scoping is important because:
- It gives direction to the AI project
- It avoids solving the wrong problem
- It saves time, money, and effort
📌 If the problem is not clear:
- The AI system may give useless results
- The project may fail completely
🔹 Example
Problem: “Farmers are facing crop loss.”
If not scoped properly:
- AI may predict weather only
But the real problem might be: - Soil health
- Pest attacks
- Lack of irrigation
👉 Correct problem scoping ensures the right AI solution.
3️⃣ IMPORTANCE OF CORRECT AND SUFFICIENT DATA
🔹 Why Data Is Important in AI
Data is the foundation of AI.
AI systems:
- Learn from data
- Make decisions using data
- Improve with more data
📌 No data = No AI
🔹 Correct Data
Correct data means:
- Accurate data
- Relevant to the problem
- Up-to-date information
Wrong data leads to:
- Wrong predictions
- Biased decisions
- Unsafe AI systems
🔹 Sufficient Data
Sufficient data means:
- Enough quantity of data
- Covers different situations
📌 Too little data:
- AI cannot learn properly
📌 Too much but irrelevant data:
- Confuses the AI system
4️⃣ ROLE OF DATA EXPLORATION BEFORE BUILDING AI MODELS
🔹 What Is Data Exploration?
Data exploration means:
- Studying collected data
- Understanding its structure
- Finding patterns and trends
🔹 Why Data Exploration Is Necessary
Before building AI models, data must be explored because:
- It helps detect missing values
- It removes unwanted data
- It improves accuracy of AI models
📌 Skipping data exploration can cause:
- Poor performance
- Biased AI
- Incorrect outputs
5️⃣ DATA ACQUISITION
🔹 What Is Data?
Data is a collection of facts, figures, or information used by AI systems to learn and make decisions.
Examples:
- Words in a message
- Photos
- Sounds
- Numbers
🔹 Types of Data
1️⃣ Text Data
- Emails
- Messages
- Reviews
- Articles
📌 Used in:
- Chatbots
- Translation apps
- Sentiment analysis
2️⃣ Image Data
- Photographs
- Medical images
- Satellite images
📌 Used in:
- Face recognition
- Object detection
- Medical diagnosis
3️⃣ Audio Data
- Voice recordings
- Music
- Phone calls
📌 Used in:
- Voice assistants
- Speech recognition
- Call analysis
4️⃣ Numerical Data
- Marks
- Temperature readings
- Sales figures
📌 Used in:
- Weather prediction
- Financial analysis
- Performance tracking
🔹 Sources of Data
1️⃣ Government Portals
- Census data
- Weather data
- Health statistics
📌 Reliable and authentic source
2️⃣ Surveys
- Questionnaires
- Feedback forms
📌 Used to collect opinions and responses
3️⃣ Sensors
- Temperature sensors
- Motion sensors
- Cameras
📌 Used in smart devices and machines
4️⃣ Online Datasets
- Open-source platforms
- Educational datasets
📌 Easily accessible for AI projects
6️⃣ DATA EXPLORATION
🔹 Cleaning Data
Data cleaning means:
- Removing duplicate data
- Filling missing values
- Correcting errors
📌 Clean data improves AI accuracy.
🔹 Understanding Patterns
Pattern recognition includes:
- Finding trends
- Grouping similar data
- Identifying relationships
📌 Patterns help AI make predictions.
🔹 Identifying Errors or Bias in Data
Bias means:
- Data favors one group over another
Errors include:
- Wrong values
- Incomplete information
📌 Identifying bias is important to build fair AI systems.
7️⃣ TYPES OF AI SYSTEMS
🔹 Rule-Based AI
Rule-Based AI works on:
- Pre-defined rules
- “If–Then” conditions
📌 Example:
- If temperature > 38°C → Alert
🔹 Characteristics:
- Simple
- No learning ability
- Works only within rules
🔹 Learning-Based AI
Learning-Based AI:
- Learns from data
- Improves over time
- Uses patterns instead of fixed rules
📌 Example:
- Spam email detection
- Recommendation systems
🔹 Characteristics:
- Flexible
- Intelligent
- Can handle complex problems
🔹 Comparison Example
| Rule-Based AI | Learning-Based AI |
|---|---|
| Uses fixed rules | Learns from data |
| No improvement | Improves with experience |
| Limited tasks | Complex tasks |
| Example: Calculator | Example: Voice assistant |
8️⃣ SUGGESTED ACTIVITIES (EXPLAINED)
🔹 Collecting Datasets from Government Websites
- Students explore public data portals
- Identify useful datasets for AI projects
🔹 Hands-On Practice of Data Collection & Exploration
- Collect sample data
- Clean and analyze it
🔹 Rule-Based vs Learning-Based AI Demonstration
- Compare outputs
- Discuss advantages and limitations
🔹 AI Project Cycle Quiz
- Questions from all stages
- Concept-based assessment
CHAPTER 3- Ethics in Artificial Intelligence
1️⃣ ETHICS – BASIC MEANING
🔹 What Are Ethics?
Ethics are a set of moral values and principles that guide human behavior and help us decide:
- What is right
- What is wrong
- What is fair
- What is unfair
Ethics help people:
- Take responsible decisions
- Treat others with respect
- Live peacefully in society
📌 Ethics are not written laws, but moral rules followed by individuals and society.
🔹 Why Are Ethics Important in Human Decisions?
Ethics are important because they:
- Guide our actions and choices
- Prevent harm to others
- Promote honesty, fairness, and justice
📌 Without ethics:
- Society would face chaos
- People could misuse power and technology
Example:
Telling the truth, helping others, respecting privacy—all are ethical actions.
2️⃣ AI ETHICS
🔹 Meaning of AI Ethics
AI Ethics refers to the moral principles and rules that guide:
- The design
- The development
- The use
of Artificial Intelligence systems.
AI ethics ensures that AI:
- Is safe
- Is fair
- Respects human values
- Does not harm individuals or society
🔹 Difference Between Human Ethics and AI Ethics
| Human Ethics | AI Ethics |
|---|---|
| Followed by humans | Applied to AI systems |
| Based on emotions and values | Based on data and programming |
| Humans take responsibility | Humans are responsible for AI actions |
| Flexible thinking | Works within given rules |
📌 Important Point:
AI does not think on its own. Humans decide the ethics for AI.
3️⃣ WHY AI ETHICS MATTER
AI systems influence:
- Education
- Healthcare
- Employment
- Security
- Social media
Therefore, ethical AI use is extremely important.
🔹 Bias in AI Systems
Bias means:
- Favoring one group over another unfairly
AI bias occurs when:
- Training data is incomplete
- Data represents only one group
- Human bias enters the system
📌 Example:
If an AI system is trained mostly on data from one region, it may give wrong results for other regions.
🔹 Privacy and Data Security
AI systems collect and use large amounts of personal data.
Ethical concerns include:
- Data misuse
- Data theft
- Unauthorized access
📌 Example:
Using someone’s personal data without permission is unethical.
🔹 Fairness and Inclusiveness
Ethical AI should:
- Treat all people equally
- Work for all genders, communities, and age groups
- Avoid discrimination
📌 Fair AI promotes equal opportunities.
🔹 Accountability in AI Decisions
Accountability means:
- Taking responsibility for AI decisions
Important questions:
- Who is responsible if AI makes a mistake?
- Who will fix the problem?
📌 Humans, not machines, must be accountable.
4️⃣ RESPONSIBLE USE OF AI
🔹 Safe and Ethical Use of AI Tools
Responsible AI use means:
- Using AI for positive purposes
- Avoiding harm
- Following rules and guidelines
Students should:
- Not misuse AI tools
- Respect privacy
- Use AI ethically in studies and daily life
🔹 Impact of Unethical AI on Society
Unethical AI can:
- Spread false information
- Increase inequality
- Violate privacy
- Cause job loss unfairly
📌 Ethical AI protects society from harm.
5️⃣ SUGGESTED ACTIVITIES (EXPLAINED)
🔹 Classroom Debates on Ethical AI Use
- Students discuss benefits and risks of AI
- Develop critical thinking skills
🔹 Balloon Debate on AI Ethical Dilemmas
- Each student represents an AI issue
- Students argue why their issue is important
- Encourages ethical reasoning
🔹 Case Discussions on Biased AI Systems
- Real-life examples of biased AI
- Identify problems and solutions
UNIT–2: Project 0: Presentation
CHAPTER 4- Domains of Artificial Intelligence
1️⃣ DOMAINS OF ARTIFICIAL INTELLIGENCE
Artificial Intelligence is divided into different domains based on the type of data used and the nature of the problem solved.
Understanding AI domains helps us:
- Choose the correct AI approach
- Build accurate AI solutions
- Avoid wrong or ineffective models
There are three main AI domains at this level:
- Data Domain
- Computer Vision Domain
- Natural Language Processing (NLP) Domain
2️⃣ DATA DOMAIN
🔹 Meaning of Data Domain
The Data Domain deals with:
- Numerical data
- Structured data
- Patterns and trends in data
In this domain, AI systems:
- Analyze numbers
- Find relationships
- Make predictions
🔹 Types of Data Used
- Marks of students
- Weather readings
- Sales records
- Health statistics
🔹 Applications of Data Domain
- Weather prediction
- Disease risk analysis
- Performance evaluation
- Market trend analysis
📌 Key Idea:
If the problem mainly involves numbers and statistics, it belongs to the Data Domain.
3️⃣ COMPUTER VISION DOMAIN
🔹 Meaning of Computer Vision
Computer Vision is the domain of AI that enables machines to:
- See
- Understand
- Interpret images and videos
Just like human eyes and brain work together, AI systems use:
- Cameras
- Image data
- Algorithms
to understand visual information.
🔹 Types of Data Used
- Images
- Videos
- Live camera feeds
🔹 Applications of Computer Vision
- Face recognition
- Traffic monitoring
- Object detection
- Medical image analysis
📌 Key Idea:
If the problem involves images or videos, it belongs to the Computer Vision Domain.
4️⃣ NATURAL LANGUAGE PROCESSING (NLP) DOMAIN
🔹 Meaning of NLP
Natural Language Processing (NLP) is the AI domain that helps machines:
- Understand human language
- Read text
- Listen to speech
- Respond meaningfully
NLP allows humans to communicate with machines naturally.
🔹 Types of Data Used
- Text messages
- Emails
- Voice commands
- Documents
🔹 Applications of NLP
- Chatbots
- Voice assistants
- Language translation
- Sentiment analysis
📌 Key Idea:
If the problem involves text or speech, it belongs to the NLP Domain.
5️⃣ MAPPING AI SOLUTIONS TO AI DOMAINS
🔹 What Does Mapping Mean?
Mapping means:
- Identifying the problem
- Matching it with the correct AI domain
Correct mapping ensures:
- Better accuracy
- Efficient AI solution
- Proper use of resources
🔹 Identifying the Correct AI Domain for a Problem
Ask these questions:
- Does the problem involve numbers? → Data Domain
- Does it involve images or videos? → Computer Vision
- Does it involve text or speech? → NLP
📌 Some problems may use more than one domain.
🔹 How Applications Use Different Domains
Many AI applications combine domains.
Example:
Voice Assistant:
- NLP → Understand voice command
- Data Domain → Process information
6️⃣ AI-BASED APPLICATIONS (REAL-WORLD EXAMPLES)
🔹 Example 1: Face Unlock in Mobile Phones
- Domain: Computer Vision
- Data: Images of faces
🔹 Example 2: Online Exam Result Analysis
- Domain: Data Domain
- Data: Marks and scores
🔹 Example 3: Chatbots on Websites
- Domain: NLP
- Data: Text and speech
🔹 Example 4: Traffic Signal Monitoring
- Domain: Computer Vision + Data Domain
- Data: Video feeds and traffic data
7️⃣ MATCHING PROBLEMS WITH AI DOMAINS
| Problem | AI Domain |
|---|---|
| Predicting rainfall | Data Domain |
| Detecting faces | Computer Vision |
| Translating languages | NLP |
| Reading handwritten text | Computer Vision + NLP |
📌 Correct matching is a key learning outcome of this unit.
8️⃣ SUGGESTED ACTIVITIES (EXPLAINED)
🔹 Student Presentations
- Students explain AI domains
- Use real-life examples
- Improve communication skills
🔹 AI Application Matching Exercises
- Match problems with correct AI domains
- Identify single or multiple domain usage
🔹 Group-Based AI Solution Discussions
- Students work in groups
- Identify problem → select domain → explain solution
CHAPTER 5- AI Use Cases and Project Cycle Applications
1️⃣ AI USE CASE ANALYSIS
🔹 What Is an AI Use Case?
An AI use case is a real-life situation or problem where Artificial Intelligence can be applied to:
- Improve efficiency
- Save time
- Increase accuracy
- Support decision-making
📌 In simple words:
An AI use case explains how AI is used to solve a specific real-world problem.
🔹 Understanding a Real-World AI Use Case
To understand an AI use case, we must know:
- What problem exists
- Who is affected
- How AI can help
- What data is required
- What outcome is expected
📌 Example:
Predicting crop disease using AI is an AI use case in agriculture.
2️⃣ BREAKING THE USE CASE INTO AI PROJECT STAGES
Every AI use case follows the AI Project Cycle.
Let us understand how a real-world problem is divided into stages.
3️⃣ MAPPING AI USE CASE TO AI PROJECT CYCLE
🔸 Stage 1: Problem Identification
In this stage:
- The real problem is clearly identified
- The goal of the AI solution is defined
📌 Questions asked:
- What is the problem?
- Who faces it?
- Why is it important to solve?
Example:
Farmers suffer losses due to crop disease.
🔸 Stage 2: Data Collection
In this stage:
- Relevant data is collected
- Data may come from:
- Sensors
- Records
- Surveys
- Images
📌 Data must be:
- Accurate
- Sufficient
- Relevant
Example:
Crop images, weather data, soil data.
🔸 Stage 3: Data Exploration
Here:
- Data is cleaned
- Errors are removed
- Patterns are studied
📌 This helps in understanding:
- Trends
- Relationships
- Useful information
Example:
Identifying patterns between weather conditions and crop disease.
🔸 Stage 4: Model Idea (Modelling)
In this stage:
- An AI model is planned
- The system learns from data
- Predictions or decisions are made
📌 This is where AI learns.
Example:
AI model predicts whether crops are healthy or diseased.
🔸 Stage 5: Evaluation
In the final stage:
- AI solution is tested
- Accuracy is checked
- Improvements are suggested
📌 If results are unsatisfactory:
- The process is repeated
- Model is improved
4️⃣ SECTOR-BASED AI PROJECTS
AI is applied in many sectors.
In this unit, we focus on four important sectors.
🌾 AGRICULTURE
🔹 AI Use Case in Agriculture
- Crop disease prediction
- Soil quality analysis
- Weather forecasting
🔹 Mapping with AI Project Cycle
- Problem: Crop loss
- Data: Crop images, weather data
- Exploration: Disease patterns
- Model: Disease prediction
- Evaluation: Accuracy of prediction
📌 Benefit:
Higher crop yield and reduced losses.
🍞 FOOD PROCESSING
🔹 AI Use Case in Food Processing
- Quality checking of food
- Sorting defective items
- Monitoring food safety
🔹 Mapping with AI Project Cycle
- Problem: Poor food quality
- Data: Images of food items
- Exploration: Defect identification
- Model: Quality classification
- Evaluation: Error rate
📌 Benefit:
Improved food quality and safety.
🥛 DAIRY FARMING
🔹 AI Use Case in Dairy Farming
- Milk quality monitoring
- Health tracking of animals
- Automated feeding systems
🔹 Mapping with AI Project Cycle
- Problem: Low milk production
- Data: Animal health records
- Exploration: Health patterns
- Model: Production prediction
- Evaluation: Output improvement
📌 Benefit:
Better dairy management and productivity.
🏥 HEALTHCARE
🔹 AI Use Case in Healthcare
- Disease diagnosis
- Medical image analysis
- Patient monitoring
🔹 Mapping with AI Project Cycle
- Problem: Late disease detection
- Data: Medical reports, images
- Exploration: Symptom patterns
- Model: Diagnosis support
- Evaluation: Accuracy and reliability
📌 Benefit:
Early diagnosis and better treatment.
5️⃣ SUGGESTED ACTIVITIES (EXPLAINED)
🔹 Project-Based Learning
- Students choose a real-life problem
- Apply AI Project Cycle steps
- Develop problem-solving skills
🔹 Group AI Projects Using AI Project Cycle
- Students work in groups
- Divide tasks among members
- Encourage teamwork and creativity
🔹 Presentation of Sector-Specific AI Solutions
- Students present their AI project
- Explain problem, data, and solution
- Improve communication skills