CLASS 8 ARTIFICIAL INTELLIGENCE : UNIT 1 – AI PROJECT CYCLE

🔹 1️⃣ SHORT NOTES FOR REVISION

1.1 Recap – Sustainability and Systems Thinking

A. Sustainability

The word “Sustainability” comes from “Sustain”, which means to maintain or support.
Sustainability means using the Earth’s resources responsibly so that future generations can also use them.
Example:
- Switching off lights when not in use
- Recycling plastic
- Planting trees

B. Society and Sustainability

A society means the group of people living together.
To make the world sustainable, everyone in society must act responsibly.
Example: If only some people save water and others waste it, sustainability cannot be achieved.

C. Sustainable Development Goals (SDGs)

The United Nations (UN) created 17 goals in 2015 to make the world better.
These goals cover three main areas:
1. Economy (Money, jobs, industry)
2. Society (People, equality, education)
3. Biosphere (Earth, nature, climate)

👉 Visit: https://sdgs.un.org/goals

D. Systems Thinking

A system is a group of connected parts that work together.
Example systems:
- The Water Cycle (evaporation → condensation → rain)
- A School (students, teachers, principal, classrooms)
- Food chain (plants → deer → tiger)
Systems are shown using System Maps, which show how things affect each other.
Example: If the number of rabbits increases, foxes will also increase (as they have more food).

1.2 Understanding Projects and Project Cycles

A. What is a Project?

A project is a set of tasks done to achieve a goal within a limited time.
Example: Making a science model, planting trees in the school, etc.

B. What is a Project Cycle?

A Project Cycle is a sequence of stages that help us plan, execute, and complete a project.
It helps to finish the work in an organized way.

C. Example: Coffee Production System

Steps in making coffee:
1. Harvesting – Picking coffee beans.
2. Processing – Removing the fruit around the bean.
3. Roasting – Heating beans to develop flavor.
4. Packaging – Packing and selling.
This shows a simple project cycle in real life.

1.3 What is an AI Project Cycle?

The AI Project Cycle is a cyclical process (repeats again and again) that helps us create an Artificial Intelligence project.
It helps us:
- Understand the process better
- Build AI projects faster
- Achieve accurate results

The 6 Stages of the AI Project Cycle are:

Problem Scoping
Data Acquisition
Data Exploration
Modelling
Evaluation
Deployment

1.4 Stages of AI Project Cycle

Stage 1: Problem Scoping

In this stage, we define what problem we are trying to solve.
This is the most important step, like setting the goal before starting work.

Method: 4Ws Framework

W	Question	Example (Air Pollution Problem)
WHO	Who faces the problem?	City people, children, traffic police
WHAT	What is the problem?	Increasing air pollution
WHERE	Where does it happen?	In cities, near factories
WHY	Why is it important to solve?	To improve health and environment

Problem Statement Example:

Our city residents are facing health problems due to air pollution when travelling on roads.
An ideal solution would monitor air quality and alert people when pollution is high.

Stage 2: Data Acquisition

Collecting data that will help AI understand and solve the problem.
Types of Data:
- Textual (words or sentences)
- Numerical (numbers, tables)
- Visual (images, videos)

Sources of Data:

Primary sources: Surveys, experiments, sensors.
Secondary sources: Websites, government portals like data.gov.in

Example:
For air pollution – collect data from air sensors, weather reports, or government datasets.

Stage 3: Data Exploration

Here, data is analyzed and shown in visual forms like graphs or charts.
Helps find patterns and trends.

Example:
Bar graph showing pollution levels in different months.
Pie chart showing the percentage of pollution sources (vehicles, factories, etc.).

Activity:
Create a table of your class students – name, height, weight, hobby – and make graphs for each.

Stage 4: Modelling

In this stage, an AI model is built using the data collected.
The model learns patterns and predicts results.

Types of AI Models:

Rule-Based AI: Works with fixed rules (e.g., If it rains → take umbrella).
Learning-Based AI: Learns from examples (e.g., AI learns to identify cats by studying many cat photos).

Example:
AI learns to predict air pollution levels using data of previous days.

Stage 5: Evaluation

Here, the AI model is tested to check how well it works.
The best model is selected based on accuracy.

Example:
Compare two AI models – choose the one that gives correct predictions most of the time.

Stage 6: Deployment

The final AI model is made available for users.
Example: A mobile app that tells real-time pollution levels or gives air quality alerts.

🔁 Quick Revision Table

Stage	Main Task	Output
1. Problem Scoping	Identify the goal	Clear problem statement
2. Data Acquisition	Collect data	Dataset
3. Data Exploration	Visualize and analyze	Graphs, insights
4. Modelling	Build AI model	Trained AI system
5. Evaluation	Test accuracy	Best model selected
6. Deployment	Release solution	Ready-to-use AI app

🔹 2️⃣ MULTIPLE CHOICE QUESTIONS (MCQs) – 20 BEST QUESTIONS

1. The AI Project Cycle is a:
A. Linear process B. Cyclical process C. Random process D. One-time activity
Ans: B

2. The main purpose of the AI Project Cycle is to:
A. Complete work quickly B. Create AI projects systematically C. Avoid teamwork D. Increase cost
Ans: B

3. The first stage of the AI Project Cycle is:
A. Evaluation B. Deployment C. Problem Scoping D. Data Acquisition
Ans: C

4. The final stage of the AI Project Cycle is:
A. Modelling B. Evaluation C. Deployment D. Data Exploration
Ans: C

5. The 4Ws framework is used in which stage?
A. Data Acquisition B. Modelling C. Problem Scoping D. Evaluation
Ans: C

6. In the 4Ws framework, “Who” refers to:
A. Who benefits B. Who causes it C. Who faces the problem D. Who funds it
Ans: C

7. Data can be of which types?
A. Textual, Numerical, Visual B. Verbal, Written, Audio C. Numeric only D. Visual only
Ans: A

8. Example of Primary Data Source is:
A. Data.gov.in B. Wikipedia C. Conducting a survey D. News articles
Ans: C

9. Example of Secondary Data Source is:
A. Experiment B. Sensor readings C. Government portals D. Field notes
Ans: C

10. The purpose of Data Exploration is to:
A. Collect data B. Visualize data to find patterns C. Discard data D. Hide data
Ans: B

11. In Data Exploration, data is shown as:
A. Text only B. Graphs and charts C. Equations D. Maps only
Ans: B

12. The process of creating an AI model is called:
A. Evaluation B. Modelling C. Problem Scoping D. Data Visualization
Ans: B

13. “If it rains → take umbrella” is an example of:
A. Learning-Based AI B. Rule-Based AI C. Predictive AI D. Deep Learning
Ans: B

14. AI that learns from examples is called:
A. Rule-Based B. Learning-Based C. Manual D. Programmed
Ans: B

15. The purpose of Evaluation stage is to:
A. Test the model accuracy B. Collect data C. Draw charts D. Define problem
Ans: A

16. The Deployment stage means:
A. Making the model available for users B. Deleting old data C. Building a graph D. Revising the dataset
Ans: A

17. Example of a deployed AI system:
A. Mobile pollution alert app B. School project report C. Pie chart D. Notebook entry
Ans: A

18. The output of Problem Scoping stage is:
A. Dataset B. Problem Statement C. Graph D. Prediction
Ans: B

19. The output of Modelling stage is:
A. Trained AI Model B. Raw Data C. Table D. Survey Result
Ans: A

20. Which of the following represents the correct order of the AI Project Cycle?
A. Data → Problem → Model → Evaluate → Deploy
B. Problem → Data → Explore → Model → Evaluate → Deploy
C. Model → Problem → Data → Explore → Deploy → Evaluate
D. Evaluate → Problem → Deploy → Explore → Data
Ans: B

🔹 3️⃣ VERY SHORT ANSWER QUESTIONS (VSAQs) – 10 QUESTIONS

1. What is sustainability?
→ Using Earth’s resources responsibly for future generations.

2. Define a project.
→ A project is a series of tasks done to achieve a specific goal in limited time.

3. What is the AI Project Cycle?
→ A cyclical process of creating and improving AI solutions in six stages.

4. Name any two stages of the AI Project Cycle.
→ Problem Scoping, Data Acquisition.

5. What is the 4Ws method?
→ A framework using Who, What, Where, and Why to define problems.

6. Give one example of primary data source.
→ Surveys or experiments.

7. Give one example of secondary data source.
→ Government datasets (e.g., data.gov.in).

8. What is Data Exploration?
→ Visualizing and analyzing data to identify patterns.

9. What happens in the Evaluation stage?
→ Models are tested and compared to select the best one.

10. What is Deployment in AI?
→ Releasing the final model for user or public use.

🔹 4️⃣ SHORT ANSWER QUESTIONS (SAQs) – 10 QUESTIONS

1. Why is sustainability important in today’s world?
→ Because natural resources are limited, and responsible use ensures their availability for future generations.

2. Why is Problem Scoping considered the foundation of AI projects?
→ It defines the problem clearly, helping to set direction and avoid wasted effort later.

3. Differentiate between Primary and Secondary Data.
→ Primary data is newly collected by the researcher; secondary data already exists and is collected by others.

4. How does Data Exploration help in AI decision-making?
→ It helps identify useful trends and patterns, guiding accurate predictions.

5. What are the benefits of the AI Project Cycle?
→ It brings clarity, accuracy, and systematic completion of AI projects.

6. Define Rule-Based AI with an example.
→ AI that follows pre-set rules, e.g., IF light is red THEN stop vehicle.

7. Define Learning-Based AI with an example.
→ AI that learns from data patterns, e.g., AI identifying cats after training on images.

8. What is the purpose of the Evaluation stage?
→ To test different models, check accuracy, and choose the most reliable one.

9. Explain the importance of Deployment.
→ It ensures the final AI solution reaches users and starts solving real problems.

10. What are the six stages of the AI Project Cycle in order?
→ Problem Scoping → Data Acquisition → Data Exploration → Modelling → Evaluation → Deployment.

🔹 5️⃣ LONG ANSWER QUESTIONS (LAQs) – 5 QUESTIONS

1. Explain the six stages of the AI Project Cycle in detail.
→

Problem Scoping: Identify and define the real-world problem using the 4Ws framework.
Data Acquisition: Gather data from various reliable sources.
Data Exploration: Understand and visualize data using charts or graphs.
Modelling: Build AI models to find patterns and make predictions.
Evaluation: Test and compare models to choose the best-performing one.
Deployment: Make the model available as an app, tool, or program for user access.

2. How can AI help in achieving Sustainable Development Goals (SDGs)?
→ AI helps analyze environmental data, track pollution, optimize energy use, and improve health and education services — directly supporting SDG goals like climate action and quality education.

3. What is Systems Thinking, and how does it relate to AI?
→ Systems Thinking studies how parts of a system interact. In AI, it helps in understanding relationships between data, users, and outcomes, improving accuracy and efficiency of models.

4. Compare Rule-Based and Learning-Based AI in detail.
→

Rule-Based AI: Works on fixed rules (IF–THEN logic), easy to program but limited in flexibility.
Learning-Based AI: Learns patterns from data, adapts over time, and can handle complex problems.
Example – Traffic light (rule-based) vs Face recognition (learning-based).

5. Explain how proper Data Acquisition and Exploration influence AI model accuracy.
→ Accurate data collection ensures the AI learns correct patterns. Data Exploration reveals hidden trends, allowing better training and testing. Both stages are vital for reliable predictions.

🔹 6️⃣ SOURCE-BASED / CASE-BASED QUESTIONS – 3 SETS

🧩 Case 1: Air Quality Monitoring AI App

A city government plans to build an AI app that warns people when air quality becomes poor. The team collects pollution data from sensors, weather stations, and online reports. They use graphs to study pollution trends and then build an AI model to predict high-risk areas.

Questions:

Identify two stages of the AI Project Cycle used in this case.
Why is Data Exploration useful for this project?
Which stage will make the app available to citizens?

Answers:

Data Acquisition and Data Exploration.
It helps find pollution trends and key causes.
Deployment stage.

🧩 Case 2: Coffee Production System

Coffee production involves harvesting, processing, roasting, and packaging. Every step must be completed in sequence for quality results.

Questions:

What does this process illustrate?
How is this example similar to an AI Project Cycle?
What lesson about organization does this example teach?

Answers:

A project cycle.
Both have ordered stages leading to a goal.
Proper sequence and planning improve quality.

🧩 Case 3: Smart Farming with AI

A farmer uses drones and sensors to monitor soil moisture and crop health. The data is analyzed to predict watering schedules, and an AI app recommends fertilizer usage.

Questions:

Name any two types of data collected.
Which stage of the AI Project Cycle involves analyzing this data?
How does this project support sustainability?

Answers:

Visual (drone images) and numerical (moisture levels).
Data Exploration.
It reduces water waste and promotes efficient farming.

🔹 7️⃣ SOLVED NCERT-BASED EXERCISE QUESTIONS

What is the AI Project Cycle?
→ It is a cyclical process consisting of six stages—Problem Scoping, Data Acquisition, Data Exploration, Modelling, Evaluation, and Deployment—to build efficient AI solutions.
What are the different stages of the AI Project Cycle and their importance?
→
- Problem Scoping – defines the problem clearly.
- Data Acquisition – collects reliable data.
- Data Exploration – identifies insights and trends.
- Modelling – builds the learning system.
- Evaluation – tests performance.
- Deployment – delivers final product for use.
Why is Problem Scoping crucial for an AI project?
→ Because it sets the goal and defines boundaries for effective solution-building.
Differentiate between Rule-Based and Learning-Based AI with examples.
→ Rule-Based AI uses predefined rules (e.g., traffic lights). Learning-Based AI learns from data (e.g., recommendation systems).
What happens in the Deployment stage?
→ The model is integrated into real applications such as websites, mobile apps, or devices for end users.