G. D.

GOENKA INTERNATIONAL SCHOOL

-------------------------------------------------------------------------------------------------------------------
1 What do you mean by project cycle? What is the significance of the AI Project Cycle?
Ans. ‘AI Project Cycle’ can be described as a systematic and sequential process that involves
effective planning, organization, coordination and development of the project, starting
from the initial planning phase and progressing through execution, completion and review.
The AI Project Cycle plays a crucial role in breaking down the development of an AI project
into different stages. This division simplifies the process, enhances clarity and makes each
step precise. Achieving the project goals is the ultimate aim of a Project Cycle.
2 What are the main stages involved in an AI project life cycle? Explain each stage briefly.
Ans. The AI Project Cycle can be described as a systematic and sequential process from the initial
planning phase till the completion and review phase.
There are 5 stages of the AI Project Cycle.
(i) Problem Scoping: Problem scoping is the process of defining and understanding the
specific boundaries and details of a problem before starting an AI project. It helps us clarify
what needs to be solved and what is the best approach to solve our problem.
(ii) Data Acquisition: This stage involves gathering and collecting relevant data for the AI
project. The sources of data are databases, web scraping, social media, sensors, surveys
and APIs.
(iii) Data Exploration: Data exploration is a way to discover hidden patterns, interesting
insights and useful information from the collected data. Data exploration helps us make
important decisions and find insights that can be used to improve things.
(iv) Modelling: AI modelling refers to the process of creating a mathematical or statistical
representation of a problem. The process of AI modelling has three essential components:
data, algorithms and training.
(v) Evaluation and Deployment: Evaluation helps to understand the performance and
effectiveness of AI model after which it is made available for the real-world use.
3 What is the 4Ws Problem Canvas tool for problem scoping? Explain each of the 4Ws.
Ans. The 4Ws Problem Canvas is a tool for identifying and organizing key elements related to a
problem. The 4Ws are as follows:
• Who: The ‘Who’ block focuses on identifying the stakeholders. These stakeholders are
directly or indirectly impacted by the problem.
• What: The ‘What’ block focuses on identifying problems. The goal is to gather specific
details like relevant documents, reports, newspaper articles or expert opinions to provide
evidence.
• Where: The ‘Where’ block helps in examining the context, situation and locations related
to a specific problem. It helps answer questions such as where the issue exists and whether
it is confined to a specific area.
 G. D. GOENKA INTERNATIONAL SCHOOL
Subject – Artificial Intelligence (417)
Chapter-1-Part-B-Revisiting AI Project Cycle &
Ethical Frameworks for AI
Grade-10 Worksheet
• Why: The ‘Why’ block focuses on why it is important to solve this problem and how it will
benefit the stakeholders involved and the society in general.
4 What is the difference between supervises, unsupervised and reinforcement learning in AI?
Provide an example of each.
Ans. Supervised Learning Unsupervised Learning Reinforcement Learning
(i) Supervised learning uses (i) Unsupervised learning (i) Reinforcement learning is
labelled data. uses unlabelled data. based on feedback and
experiences.
(ii) Supervised learning is (ii) Unsupervised learning is (ii) Q-learning and SARSA
applicable in Regression and applicable in Clustering and are Reinforcement learning
Classification problems. Association problems. algorithms.
(iii) Example: Identifying (iii) Example: Finding groups (iii) Example: Super Mario
dogs based on labelled dog of similar images. video game.
images.
5 Describe the architecture of an Artificial Neural Network.
Ans. Artificial Neural Network primarily consists of three layers:
• Input Layer: The first layer is known as the input layer, responsible for receiving input
data. It does not perform any processing but rather passes the data to the next layer.
• Hidden Layer: The second layer, known as the hidden layer, is where the actual
computation takes place. The output generated by each node in the hidden layer is then
passed on to the next hidden layer, if present, or to the final layer.
• Output Layer: The output layer provides the final processed result or prediction based on
the computations performed in the hidden layer.