OVERVIEW OF

ARTIFICIAL
INTELLIGENCE
CHAPTER
1
 INTRODUCTION TO ARTIFICIAL INTELLIGENCE

• Definition of Artificial Intelligence (AI):

• AI can be broadly defined as the capability of a machine to imitate intelligent
human behavior, encompassing tasks such as learning, reasoning, problem-
solving, language understanding, and even decision-making.
 Can machines
Thinking: think?
Thinking is the facility to reason, analyze, evaluate, and formulate ideas and concepts.
Not every being capable of thinking is intelligent.
Intelligence:
Intelligence is perhaps akin to efficient and effective thinking.

Guess the next number

1,3,6,10,15,21 ?

1,2,2,3,3,3,4,4,4,4 ?

2,3,3,5,5,5,7,7,7,7 ?
Statement 1: Computers are made of silicon and power supplies, and therefore are not capable of thinking,”

Statement 2: “Computers perform much faster than humans and therefore must be more intelligent than
humans.”

The truth is most likely somewhere between these two extremes.

Artificial Intelligence is the science of making machines do things that

would require intelligence if done by man.

Raphael-
 IQ
Test
Question: If you rearrange the letters "CIFAIPC" you get the name of
a/an:

•Answer: Ocean (PACIFIC).

Question:What comes next in the sequence: 2, 6, 12, 20, 30, ...?

Answer: 42 (The pattern is n2+nn^2 + nn2+n).

Question:What is 45% of 200?

Answer: 90.
 IQ
Test
Question:What number completes the sequence: 1, 4, 9, 16, 25, ...?

Answer: 36 (These are squares of natural numbers: 12,22,32,42,52,621^2, 2^2, 3^2, 4^2, 5^2,
6^212,22,32,42,52,62).

Question:Which shape does not belong: Circle,Triangle, Square, Rectangle?

Answer: Circle (It has no corners, while the others do).

• Importance of AI in modern technology:
• Enhanced efficiency and Productivity.
• Improved decision making.
• Innovation in Autonomous Systems.
• Personalization and Customer Experience.

• AI applications in various fields:

• Healthcare
• Finance
• Manufacturing
• Transportation.
 THE TURING TEST

• Definition and purpose of the Turing

Test:
 The Turing Test, proposed by the British mathematician and computer scientist Alan Turing in 1950,
is a measure of a machine's ability to exhibit intelligent behavior that is indistinguishable from that
of a human.
 It has been a foundational concept in the field of artificial intelligence (AI) and
has started numerous discussions and advancements in AI research.

Alan
Turing
 Turing
Test
• What is your favourite book and why?

• D o you prefer cats or dogs? Explain your

choice.

• What are the current weather conditions?

• Are you afraid of dying?

• How does the dark make you feel?

• How does it feel like to be in love?

 Criticism s on Turing Test
Block’s Criticism:
Block presents a hypothetical machine, called "Blockhead," which is pre-programmed with all
possible
conversational responses for a Turing Test scenario. This machine could pass the Turing Test by simply retrieving
pre-stored responses, without any understanding or cognitive processes. Block uses this to show that the Turing
Test can be fooled by a machine that lacks genuine intelligence.

Question) Are you afraid of

dying?
Searle’s Criticism
John Searle’s criticism of the Turing test is more fundamental. Imagine an interrogator who
asks questions as expected—this time, however, in Chinese. In a separate room is someone who does not
know Chinese, but does have a detailed rulebook.Although the Chinese questions appear as a series of
squiggles,
the person in the room consults the rulebook, processes the Chinese characters according to the rules,
and responds with answers written using Chinese characters.

In the same way, a computer running a program receives,

processes, and responds with symbols without learning or
understanding what the symbols themselves mean.
 STRO N G AI VS WEAK
AI
• W eak AI:
Weak AI (also known as Narrow AI) is a type of artificial intelligence designed and trained to
perform specific tasks. it is focused on particular functions within defined parameters.
 Task specific
 Lack of consciousness and understanding.
 Dependence on Pre-defined data.
Examples:
Siri, Alexa, Amazon, Netflix, Self driving cars.
• Strong
AI: Strong AI, also known as Artificial General Intelligence (AGI), is a type of artificial intelligence that aims
to replicate human cognitive abilities across a wide range of tasks. Unlike weak AI, which is designed for specific
tasks, strong AI seeks to perform any intellectual task that a human can.
 General Intelligence.
 Consciousness and Self Awareness.
 Learning and Adaption.
 Human Like reasoning.
AI Types
Artificial
Intelligence

Artificial General
Narrow AI/Weak AI Intelligence/Strong Super AI
AI

Reactive Machine AI Theory of Surpass Human

Ex: Deep Blue Super
Mind AI. Intelligence
Computer.

Limited Memory AI
Ex: GPTs,
Generative AI
Chat Bots.
 H EURISTIC S IN A I

Definition of
heuristics:
 A heuristic is a rule of thumb for solving a problem. In other words, a heuristic is a
set of guidelines that often works to solve a problem.
 With a heuristic, a favorable outcome is likely, but is not guaranteed.
 AI applications often rely on the application of heuristics. Contrast a heuristic with
an algorithm, which is a prescribed set of rules to solve a problem and whose
output is entirely predictable.
 Heuristics in problem-solving:
George Polya’s Water Jug Problem:
You are provided with two jugs of sizes m and n respectively; and you are required to measure r liters of water
where m, n, and r are all different quantities.An instance of this problem is:
 How can you measure exactly twelve liters of water from a tap or a well when you have only an eight-liters
jug and an eighteen liters jug?

a) Initial b) Final
State State
To actually solve this problem, you reverse the order of the
states.:
d) First fill the 18-liter pail.
Then fill and empty the eight-liter pail twice by transferring water from the 18-liter pail.
c) This leaves you with two liters in the 18-liter pail .
b) Pour the last two liters into the eight-liter pail.
a) Fill the 18-liter pail again from the tap or well, and pour water from the larger container to fill the
eight-liter pail, which removes six liters from the 18, leaving 12 liters in the larger pail (state a).
Examples of heuristic methods used in
AI:
1) Greedy Algorithms
Example: D ijkstra's Algorithm

2) Genetic Algorithms
Example:Traveling Salesman
Problem

3) A* search algorithm
Example: Path finding in video
games

4) Ant colony optimization

Example:Vehicle Routing Problem
5) Hill Climbing
Example: Chessboard
 IDENTIFYING PROBLEMS SUITABLE FOR
AI
What makes a problem suitable for A I?
There are three characteristics that are common to most AI problems:
1.AI problems tend to be large.
2.They are computationally complex and cannot be solved by straightforward
algorithms.
3.AI problems and their domains tend to embody a large amount of human expertise.
Some types of problems are better solved using AI, whereas others are more suitable
for traditional computer science approaches involving simple decision-making or exact
computations to produce solutions. Let us consider a few examples:

• Medical diagnosis
• Shopping using a cash register with barcode scanning
• ATMs
• Two person games such as chess and checkers
Examples problems of well suited for
fo

AI:
1) Medical Image Processing: Detecting diseases from medical images such as X-rays, MRIs, and CT scans.

2) Natural Language Processing: Language translation, sentiment analysis, and chatbot development.

3) Recommender Systems: Personalized content recommendations on streaming platforms (Netflix, Spotify)

and ecommerce sites (Amazon).

4) Autonomous Vehicles: Self driving cars and drones.

5) Fraud detection: Identifying fraudulent transactions in financial systems.

6) Supply chain optimization: Optimizing routes and inventory levels in logistics and supply chain
management.
 Activity
Identify the tasks suitable for AI.

1. Tasks Requiring Deep Empathy and Emotional Intelligence.

2. Creative Processes

3. Predictive Maintenance

4. Understanding and Interpreting Complex Social and Cultural

Contexts.

5. Personalized Care and Human Interaction.

6. Supply C hain Optimization.

7. Cybersecurity.
 Activity: Identifying Problems Suitable for
AI
Topics:
1) A hospital wants to improve its diagnostic accuracy for detecting early-stage cancer using patient data and
medical imaging.
2) A large retail chain aims to optimize its inventory levels and reduce waste by predicting product
demand across different stores.
3) A telecommunications company plans to enhance its customer service by implementing AI chatbots to
handle common queries and issues.
4) A bank seeks to reduce fraudulent transactions by implementing an AI system that can detect
unusual patterns and behaviors in real-time.
5) An automotive manufacturer is developing self-driving cars and needs to ensure that the AI can
navigate
complex urban environments safely.
Discuss:
What is the problem described in the case study?

Can AI be applied to solve this problem? Why or why not?

If AI is suitable, what type of AI solution would you propose (e.g., machine learning,

natural language processing, computer vision)?

What are the potential benefits and challenges of applying AI to this problem?
 APPLICATIONS A N D METHODS OF
AI

A I applications in various
fields:
• Healthcare
• Finance
• Manufacturing
• Transportation.
Healthcare:
Continuous
D ata Learning
Collection and
Feedback

Treatmen
Data
t
Processing
Planning

Model Diagnosis
Training
Finance:
Customer
Service/Continu
D ata
o us Learning
Collection
and Feedback

Algorithmi
Data
c
Processing
Trading

Model Fraud
Training D etection
Manufacturing:
Continuous
O ptimization
Learning
Data of Supply
and
Collection C hain
Feedback
Management

Q uality
D ata
control
Processing
and
inspection

Predictive
Model
Maintenanc
Training
e
 Autonomous
Transportation Vehicles

Data
Collectio
n Traffic
Managemen
t
Data
Processin
g
Predictive
Maintenance
Model
Trainin
g

Drone Delivery
 AI Methods
Machine Learning

 Deep Learning

 Natural Language Processing

 Computer Vision

 Expert Systems

 Fuzzy Logic

 Evolutionary Algorithms
Computer
Vision:
Computer Vision enables machines to interpret and make decisions based on visual
data.
Image Classification
• Description: Assigning a label to an entire image based on its content.
• Applications: Object recognition, medical imaging diagnostics.
Object Detection
• Description: Identifying and locating objects within an image.
• Applications: Surveillance, autonomous vehicles, facial recognition.
Image Segmentation
• Description: Partitioning an image into multiple segments (regions or objects).
• Applications: Medical image analysis, scene understanding.
Expert Systems

Expert Systems emulate the decision-making ability of a human expert.

•Rule-Based Systems
• Description: Use predefined rules to analyze information and make decisions.
• Applications: Diagnostic systems, troubleshooting guides.
•Inference Engines
• Description: Apply logical rules to the knowledge base to deduce new
information.
• Applications: Medical diagnosis, financial forecasting.
Fuzzy Logic:

Fuzzy Logic deals with reasoning that is approximate rather than fixed and exact.

•Description: Uses degrees of truth rather than the usual true/false or 0/1 binary
logic.

• Applications: Control systems, decision-making systems, pattern

recognition.
Evolutionary algorithms:
Evolutionary Algorithms are optimization algorithms inspired by natural evolution.

Genetic Algorithms
• Description: Search heuristics that mimic the process of natural evolution.
• Applications: Optimization problems, machine learning.
Swarm Intelligence
• Description: Inspired by the collective behavior of decentralized, self-organized
systems.
• Examples: Ant colony optimization, particle swarm optimization.
• Applications: Optimization, robotics, network design.
 Assignment

EARLY HISTO RY OF
AI
• Milestones in the early development of AI (1950s-1970s)
• Key figures and their contributions (e.g., John McCarthy, Marvin
Minsky)
• Notable early AI projects and breakthroughs

Note: Use any AI

tool.
 Assignment

RECENT HISTO RY OF
AI
• Developments in AI from the 1980s to the present
• Advances in machine learning and deep learning
• Major AI research achievements and technological
advancements

Note: Use any AI

tool.
 Assignment

AI IN THE NE W
MILLENNIUM
• The current state of AI research and development
• Emerging trends and future directions in AI
• The impact of AI on society and potential ethical
considerations

Note: Use any AI

tool.
E N D OF C H A P T E R
1