IMPLEMENTATION

OF
AI AND ML USING PYTHON

Dr Ramesh Kumar Mojjada

Assistant Professor
Department of ECM
KL University
1
 ARTIFICIAL INTELLIGENCE (AI)

 Artificial refers to something which is made by human or

non natural thing and Intelligence means ability to
understand or think.

 Artificial Intelligence is a technique which enables

machine to mimic human behavior

 Artificial Intelligence is the branch of computer science

by which we can create intelligent machines which can
behave like a human, think like humans, and able to make
decisions
 APPROACHES OF AI
 There are four approaches of AI

 1) Acting humanly (The Turing test approach)

========================================

 The Turing test, created and developed by Alan Turing

during the ’50s, is the most famous test related to
Artificial Intelligence capabilities.

 The objective of the Turing test is to test the machine’s

ability to express intelligent behavior indistinguishable
from the one of a human.
 THE TURING TEST APPROACH

 Participants
 Actor A : A human evaluator
 Actor B : Another human
 Actor C : The machine you want to test

 Setting
 The participants are separated
 The conversation has to take place on text-only channels,
like screen and keyboard
 The evaluator knows in advance that either (B) or (C) is a
machine, but he does not know which one
 THE TURING TEST APPROACH

 The Test

 The evaluator (A) can have a conversation with (B) or (C)

 The evaluator has to reliably tell the machine from the

human. If he can’t, the machine has passed the test!
 APPROACHES OF AI

 2) Thinking humanly (The cognitive modelling approach)

===============================================

 Cognitive sciences

Computer
Human system
Brain
 APPROACHES OF AI

 3) Acting rationally (Rational agent approach)

===============================================

 Doing / behaving rightly

 Generalized approach

 Maximizing expected performance

 APPROACHES OF AI

 4) Thinking rationally ( Law of thought approach)

===============================================

 Right thinking

 Requires 100 % knowledge

 Too many computations are required

 MACHINE LEARNING

Machine learning is one of the most important branches of

Artificial intelligence . Machine Learning is a subset of
Artificial Intelligence

 Machine Learning Definition

 =============================
 Definition by Arthur Samuel(1959)
 Machine Learning is the field of study that gives the
computers the ability to learn without being explicitly
programmed

Itis an application of AI that provide system the ability to

automatically learn and improve from experience.
 MACHINE LEARNING

Definition by Tom Mitchell(1998):

===============================================
A computer program is said to learn from experience E with
respect to some task T and some performance measure P , if
its performance on T , as measured by p , improves with
experience E .

Example : email program watched which emails you do or

do not mark as spam , and based on that learns how to
better filter spam.
 MACHINE LEARNING

 Classifying email as spam or not spam => Task


watching you label emails as spam or not spam =>
Experience

 The number( or fraction) of emails correctly classified as

spam / not spam => performance measure
 MACHINE LEARNING

 Classifying email as spam or not spam => Task

watching you label emails as spam or not spam =>

Experience

 The number( or fraction) of emails correctly classified as

spam / not spam => performance measure
APPLICATIONS OF AI AND ML
CLASSIC/NON-ADAPTIVE MACHINES
 ADAPTIVE MACHINES

continuous ability to adapt its behavior to external

signals ( datasets or real-time inputs ) and , like a
human being, to predict the future using uncertain
and fragmentary pieces of information.
TRADITIONAL PROGRAMMING VS
MACHINE LEARNING
Traditional Programming

Input Data Output

Program
TRADITIONAL PROGRAMMING VS
MACHINE LEARNING
Machine Learning

Input Data Program

Output label
 WHEN DO WE USE MACHINE LEARNING

ML is used when:

Human expertise does not exist (navigating on Mars)

Humans can’t explain their expertise (speech

recognition)

 Models must be customized (personalized medicine)

Models are based on huge amounts of data (genomics)

 WHEN DO WE USE MACHINE LEARNING

Learning is not always useful:

There is no need to “learn” to calculate payroll

A classic example of a task that requires machine learning

is recognizing handwritten digits
 SAMPLE APPLICATIONS

Web search

Computational biology

Finance

E-commerce

Space exploration
 SAMPLE APPLICATIONS

Space exploration

Robotics

Information extraction

Social networks

Debugging software
 MACHINE LEARNING PARADIGMS

Supervised Learning
=========================================
Learn an input and output map

Regression: continuous output

Classification : categorical output

 MACHINE LEARNING PARADIGMS

Unsupervised Learning
======================
Discover patterns in the data

Clustering : cohesive grouping

Association : frequent co-occurrence

 MACHINE LEARNING PARADIGMS

Reinforcement Learning
======================
Reinforcement learning is a type of machine learning where
an agent learns to behave in an environment by performing
actions and observing the results.
 BUILDING AI AND ML APPLICATOINS
USING PYTHON

Why Python to build AI and ML applications ?

Python is simple to learn and easy to use.

Built in library support to develop any type of applications.

concise code.
 S ?
E R IE
QU
A NY
 U !
YO
N K
HA
T