Class IX Artificial Intelligence: Detailed Notes (Syllabus Till June)
Unit 1: AI Reflection, Project Cycle and Ethics
1.1 Understanding AI: Domains and Applications
• Define Artificial Intelligence (AI):
o Artificial Intelligence (AI) is a branch of computer science that focuses
on creating machines or computer systems capable of performing
tasks that typically require human intelligence.
o These tasks include learning, problem-solving, decision-making,
understanding language, recognizing patterns, and perceiving the
world.
o The goal of AI is to enable machines to think and behave in ways that,
if observed in humans, would be considered intelligent.
• Applications of AI in Everyday Life:
o Voice Assistants: (e.g., Siri, Alexa, Google Assistant) help us interact
with devices using voice commands.
o Recommendation Systems: (e.g., Netflix, YouTube, Amazon) suggest
movies, videos, or products based on our past preferences and
behavior.
o Facial Recognition: Used in smartphones for unlocking, and in
security systems for identification.
o Self-driving Cars: Use AI to perceive their surroundings, navigate, and
make driving decisions.
o Chatbots: Provide automated customer service, answering questions
and resolving queries online.
o Spam Filters: Identify and block unwanted emails from our inboxes.
o Language Translation: (e.g., Google Translate) translates text or
speech between different languages.
• The Three Domains of AI and Their Applications:
1. Computer Vision (CV):
Focus: Enables machines to "see," interpret, and understand
visual information from images and videos.
Applications: Object detection (recognizing items), facial
recognition, image classification, medical image analysis, quality
control in manufacturing.
2. Natural Language Processing (NLP):
Focus: Enables machines to understand, interpret, and
generate human language (both spoken and written text).
Applications: Voice assistants, language translation, chatbots,
sentiment analysis, text summarization, spam detection.
3. Data Science (DS):
Focus: Involves extracting knowledge and insights from large
and complex datasets using scientific methods, processes, and
algorithms.
Applications: Recommendation engines, fraud detection,
predictive analytics (forecasting trends), targeted advertising,
market research.
1.2 The AI Project Cycle - II The AI Project Cycle is a systematic approach to
developing AI solutions. It provides a structured roadmap from understanding a
problem to deploying the solution.
• Importance of the AI Project Cycle:
o Provides a structured and systematic way to build AI solutions.
o Ensures that all necessary steps are followed, reducing errors and
increasing efficiency.
o Helps in clearly defining the problem and understanding its scope.
o Guides the collection, preparation, and analysis of data.
o Facilitates effective model building, evaluation, and deployment.
o Leads to more robust, effective, and ethical AI solutions.
• To Structure the AI Problem Statement with the AI Project Cycle (4Ws
Problem Canvas):
o Problem Scoping is the first stage of the AI Project Cycle, where you
use the "4 Ws" framework to define the problem:
Who: Who are the stakeholders or the people affected by this
problem? Who will benefit from the AI solution?
What: What is the actual problem that needs to be solved?
What kind of data will be needed? What is the expected
outcome?
Where: Where does this problem occur? What is the context or
environment of the problem?
Why: Why is it important to solve this problem? What is the real-
world impact or benefit of finding a solution?
o Structuring the problem statement using these 4Ws helps ensure a
clear understanding of the project's goals and scope before
proceeding.
1.3 AI Ethics - II
• The Difference Between Ethics and Morality:
o Morality: Refers to an individual's own principles regarding right and
wrong. It's often personal, shaped by upbringing, culture, and
experiences. Morality is about an individual's beliefs.
Example: An individual might believe it is morally wrong to lie,
regardless of the situation.
o Ethics: Refers to a set of rules, principles, or standards of conduct for
a group or profession, designed to guide behavior and decision-making
towards what is good or right. Ethics are often formalized and shared
within a community or organization.
Example: A doctor's code of ethics requires them to prioritize
patient well-being, even if it conflicts with their personal moral
beliefs in some cases.
o Importance in AI: While individuals involved in AI development have
their own morals, ethical frameworks provide a shared set of guidelines
for the entire AI community to ensure responsible and beneficial AI
creation, especially given the broad impact of AI.
• The Ethical Scenarios Faced While Building AI Solutions:
o When building AI, various ethical dilemmas can arise:
Bias and Fairness: The AI might produce unfair or
discriminatory results due to biased training data (e.g., an AI for
loan approvals unfairly rejecting certain groups).
Privacy Violations: AI often needs a lot of data, which can
include sensitive personal information, leading to concerns
about data security and misuse.
Transparency and Explainability: It can be hard to understand
how an AI makes a decision ("black box" problem), making it
difficult to trust or audit its behavior.
Accountability: If an AI system makes an error or causes harm,
it's challenging to determine who is responsible (e.g., the
developer, the user, the company).
Job Displacement: AI's ability to automate tasks can lead to job
losses in certain industries.
Misinformation and Malicious Use: AI can be used to create
fake content (deepfakes) or for surveillance that infringes on
rights.
• AI Bias and to Identify Bias in AI:
o AI Bias: This occurs when an AI system produces results that are
systematically unfair or prejudiced towards particular individuals or
groups. It often reflects and amplifies existing societal biases.
o How to Identify Bias:
Review Training Data: Check if the data used to train the AI is
diverse and representative, or if it lacks data from certain
groups.
Analyze Outputs: Test the AI system with different inputs (e.g.,
images of people from various backgrounds, different linguistic
styles) and carefully observe if the outcomes are equitable or if
they show favoritism or discrimination.
Seek Diverse Perspectives: Involve people from various
backgrounds in the testing and evaluation process to spot
biases that might otherwise be overlooked.
Fairness Metrics: Use specific technical metrics designed to
measure fairness in AI models.