Introduction to ML

Chapter 1

Overview

1
Introduction
Machine learning is a growing technology which enables
computers to learn automatically from past data.
Machine learning uses various algorithms for building
mathematical models and making predictions using
historical data or information.
Currently, it is being used for various tasks such as image
recognition, speech recognition, Email filtering, Facebook
auto-tagging, recommender system, and many more.

2
 What is Machine Learning?
In the real world, we are surrounded by humans who
can learn everything from their experiences with their
learning capability, and we have computers or
machines which work on our instructions.
But can a machine also learn from experiences or past
data like a human does? So here comes the role of
Machine Learning.

3
 What is Machine Learning…Cont’D
Machine Learning is said to be a subset of artificial
intelligence that is mainly concerned with the
development of algorithms which allow a computer to
learn from the data and past experiences on their own.
The term machine learning was first introduced by
Arthur Samuel in 1959.
Arthur Samuel, an early American leader in the field of
computer gaming and artificial intelligence, coined the
term “Machine Learning” in 1959 while at IBM.

4
 What is Machine Learning…Cont’D
He defined machine learning as “the field of study that
gives computers the ability to learn without being
explicitly programmed.”
However, there is no universally accepted definition for
machine learning. Different authors define the term
differently. Here are more definitions.
1. Tom Mitchell (1998): a computer program is said to learn
from experience E with respect to some class of tasks T and
performance measure P, if its performance at tasks in T, as
measured by P, improves with experience E.

5
 What is Machine Learning…Cont’D
Examples i) Handwriting recognition learning
problem, Task T: Recognizing and classifying handwritten
words within images, Performance P: Percent of words
correctly classified, Training experience E: A dataset of
handwritten words with given classifications
ii) A robot driving learning problem, Task T: Driving on
highways using vision sensors, Performance measure P:
Average distance traveled before an error, training
experience E: A sequence of images and steering commands
recorded while observing a human driver

6
 What is Machine Learning…Cont’D
iii) A chess learning problem, Task T: Playing chess,
Performance measure P: Percent of games won against
opponents, Training experience E: Playing practice games
against itself
2. Machine learning is programming computers to optimize
a performance criterion using example data or past
experience.
We have a model defined up to some parameters, and
learning is the execution of a computer program to optimize
the parameters of the model using the training data or past
experience.
7
 What is Machine Learning…Cont’D
The model may be predictive to make predictions in the
future, or descriptive to gain knowledge from data, or both.
[Link] field of study known as machine learning is concerned
with the question of how to construct computer programs
that automatically improve with experience.
4. A computer program which learns from experience is
called a machine learning program or simply a learning
program. Such a program is sometimes also referred to as a
learner.
5. Machine Learning is defined as the study of computer
algorithms for automatically constructing computer software
through past experience and training data.
8
 What is Machine Learning …Cont’D
6. It is a branch of Artificial Intelligence and computer
science that helps build a model based on training data and
make predictions and decisions without being constantly
programmed.
7. Machine Learning is the study of learning algorithms
using past experience and making future decisions.
8. Machine Learning is defined as the branch of Artificial
Intelligence and computer science that focuses on learning
and improving the performance of computers/machines
through past experience by using algorithms.

9
 Machine Learning
Machine learning enables a machine to automatically learn
from data, improve performance from experiences, and
predict things without being explicitly programmed.
With the help of sample historical data, which is known as
training data, machine learning algorithms build a
mathematical model that helps in making predictions or
decisions without being explicitly programmed.
Machine learning brings computer science and statistics
together for creating predictive models. It constructs or uses
the algorithms that learn from historical data.

10
 Machine Learning …Cont’D
The more we will provide the information, the higher will
be the performance. A machine has the ability to learn if it
can improve its performance by gaining more data.
Machine learning is a part of artificial intelligence and the
subfield of Data Science.
It is a growing technology that enables machines to learn
from past data and perform a given task automatically.
It allows the computers to learn from the past experiences
by its own, it uses statistical methods to improve the
performance and predict the output without being explicitly
programmed.
11
 Machine Learning …Cont’D
A)It is very hard to write programs that solve problems like
recognizing a face.
We don’t know what program to write because we don’t
know how our brain does it.
Even if we had a good idea about how to do it, the program
might be horrendously complicated.
B)Instead of writing a program by hand, we collect lots of
examples that specify the correct output for a given input.
C)A machine learning algorithm then takes these examples
and produces a program that does the job.

12
 Machine Learning …Cont’D
The program produced by the learning algorithm may look
very different from a typical hand-written program. It may
contain millions of numbers.
If we do it right, the program works for new cases as well as
the ones we trained it on.
Some more examples of tasks that are best solved by using
a learning algorithm.
A)Recognizing Patterns:
Facial identities or facial expressions
Handwritten or Spoken words
Medical images
13
 Machine Learning …Cont’D
B)Generating Patterns:
 Generating images or motion sequences
C)Recognizing Anomalies:
Unusual sequences of credit card transactions
Unusual patterns of sensor readings in a nuclear power
plant or unusual sound in your car engine.
D)Prediction:
Future stock prices or currency exchange rates
A classic example of a task that requires machine learning:
It is very hard to say what makes a 2.

14
Machine Learning …Cont’D

15
 How does Machine Learning Work?
A Machine Learning system learns from historical data,
builds the prediction models, and whenever it receives new
data, predicts the output for it.
The accuracy of predicted output depends upon the
amount of data, as the huge amount of data helps to build a
better model which predicts the output more accurately.
Suppose we have a complex problem, where we need to
perform some predictions, so instead of writing a code for it,
:

16
 How does Machine Learning Work?
We just need to feed the data to generic algorithms, and
with the help of these algorithms, machine builds the logic
as per the data and predict the output.
Machine learning has changed our way of thinking about
the problem.
The following block diagram explains the working of
Machine Learning algorithm

17
How does Machine Learning Work?

18
What is Machine Learning?
Machine Learning is defined as the branch of Artificial
Intelligence and computer science that focuses on learning
and improving the performance of computers/machines
through past experience by using algorithms.
AI is used to make intelligent machines/robots, whereas
machine learning helps those machines to train for
predicting the outcome without human intervention.

19
How does Machine Learning Work?
Machine Learning uses algorithms and techniques that
enable the machines to learn from past experience/trends
and predict the output based on that data.

20
How does Machine Learning Work?
However, machine learning access a huge amount of data
using data pre-processing. This data can be either
structured, semi-structured, or unstructured.
Further, this data is fed through some techniques and
algorithms to machines, and then based on previous trends;
it predicts the outputs automatically.

21
Features of Machine Learning:
Machine learning uses data to detect various patterns in a
given dataset.
It can learn from past data and improve automatically.
It is a data-driven technology.
Machine learning is much similar to data mining as it also
deals with the huge amount of the data.

22
Need for Machine Learning
The need for machine learning is increasing day by day.
The reason behind the need for machine learning is that it
is capable of doing tasks that are too complex for a person
to implement directly.
As a human, we have some limitations as we cannot access
the huge amount of data manually, so for this, we need
some computer systems and here comes the machine
learning to make things easy for us.

23
Need for Machine Learning …Cont’D
We can train machine learning algorithms by providing
them huge amount of data and let them explore the data,
construct the models, and predict the required output
automatically.
The performance of the machine learning algorithm
depends on the amount of data, and it can be determined by
the cost function.
With the help of machine learning, we can save both time
and money.

24
Need for Machine Learning …Cont’D
The importance of machine learning can be easily
understood by its uses cases.
Currently, machine learning is used in self-driving cars,
cyber fraud detection, face recognition, and friend
suggestion by Facebook, etc.
Various top companies such as Netflix and Amazon have
built machine learning models that are using a vast amount
of data to analyze the user interest and recommend product
accordingly.

25
 Need for Machine Learning …Cont’D
The following are some key points which show the
importance of Machine Learning:
 Rapid increment in the production of data
 Solving complex problems, which are difficult for a
human
 Decision making in various sector including finance
 Finding hidden patterns and extracting useful
information from data.

26
History of Machine Learning
Before some years (about 40-50 years), machine learning
was science fiction, but today it is the part of our daily life.
Machine learning is making our day to day life easy from
self-driving cars to Amazon virtual assistant "Alexa".
However, the idea behind machine learning is so old and
has a long history.
The following are some milestones which have occurred in
the history of machine learning:

27
History of Machine Learning …Cont’D

28
The early history of Machine Learning
(Pre-1940):
1834: In 1834, Charles Babbage, the father of the computer,
conceived a device that could be programmed with punch
cards. However, the machine was never built, but all modern
computers rely on its logical structure.
1936: In 1936, Alan Turing gave a theory that how a
machine can determine and execute a set of instructions.

29
The era of stored program computers:
1940: In 1940, the first manually operated computer,
"ENIAC" was invented, which was the first electronic
general-purpose computer.
After that stored program computer such as EDSAC in 1949
and EDVAC in 1951 were invented.
1943: In 1943, a human neural network was modeled with an
electrical circuit.
In 1950, the scientists started applying their idea to work
and analyzed how human neurons might work.

30
Computer Machinery and Intelligence:
1950: In 1950, Alan Turing published a seminal paper,
"Computer Machinery and Intelligence," on the topic of
artificial intelligence.
In his paper, he asked, "Can machines think?"

31
Machine intelligence in Games:
1952: Arthur Samuel, who was the pioneer of machine
learning, created a program that helped an IBM computer to
play a checkers game.
It performed better more it played.
1959: In 1959, the term "Machine Learning" was first
coined by Arthur Samuel.

32
The first "AI" winter:
The duration of 1974 to 1980 was the tough time for AI and
ML researchers, and this duration was called as AI winter.
In this duration, failure of machine translation occurred,
and people had reduced their interest from AI, which led to
reduced funding by the government to the researches.

33
 Machine Learning from theory to reality
1959: In 1959, the first neural network was applied to a real-world
problem to remove echoes over phone lines using an adaptive
filter.
1985: In 1985, Terry Sejnowski and Charles Rosenberg invented
a neural network NETtalk, which was able to teach itself how to
correctly pronounce 20,000 words in one week.
1997: The IBM's Deep blue intelligent computer won the chess
game against the chess expert Garry Kasparov, and it became the
first computer which had beaten a human chess expert

34
Machine Learning at 21st Century
2006: In the year 2006, computer scientist Geoffrey Hinton
has given a new name to neural net research as "deep
learning," and nowadays, it has become one of the most
trending technologies.
2012: In 2012, Google created a deep neural network which
learned to recognize the image of humans and cats in
YouTube videos.
2014: In 2014, the Chabot "Eugen Goostman" cleared the
Turing Test. It was the first Chabot who convinced the 33%
of human judges that it was not a machine.

35
Machine Learning at 21st Century
2014: DeepFace was a deep neural network created by Facebook,
and they claimed that it could recognize a person with the same
precision as a human can do.
2016: AlphaGo beat the world's number second player Lee sedol
at Go game.
In 2017 it beats the number one player of this game Ke Jie.
2017: In 2017, the Alphabet's Jigsaw team built an intelligent
system that was able to learn the online trolling. It used to read
millions of comments of different websites to learn to stop online
trolling.

36
Machine Learning at Present:
Now machine learning has got a great advancement in its
research, and it is present everywhere around us, such as self-
driving cars, Amazon Alexa, Catboats, recommender system, and
many more.
It includes Supervised, Unsupervised, and Reinforcement
learning with clustering, classification, decision tree, SVM
algorithms etc.
Modern machine learning models can be used for making
various predictions, including weather prediction, disease
prediction, stock market analysis, etc.

37
 Classification of Machine Learning:
Ata broad level, machine learning can be classified into
three types:
Supervised Learning
Unsupervised Learning
Reinforcement Learning

38
1) Supervised Learning
Supervised learning is a type of machine learning
method in which we provide sample labeled data to the
machine learning system in order to train it, and on that
basis, it predicts the output.
The system creates a model using labeled data to
understand the datasets and learn about each data,
once the training and processing are done then we test
the model by providing a sample data to check whether
it is predicting the exact output or not. 39
1) Supervised Learning
The goal of supervised learning is to map input data
with the output data.
The supervised learning is based on supervision, and it
is the same as when a student learns things in the
supervision of the teacher. The example of supervised
learning is spam filtering.
Supervised learning can be grouped further in two
categories of algorithms:
 Classification Definition: Classification is used when the output variable is a category (or class).

Goal: To predict which class or category an input belongs to.

 Regression Definition: Regression is used when the output variable is a real or continuous value.

Goal: To predict a number based on input features. 40

2) Unsupervised Learning
Unsupervisedlearning is a learning method in which a
machine learns without any supervision.
The training is provided to the machine with the set of
data that has not been labeled, classified, or
categorized, and the algorithm needs to act on that data
without any supervision.
The goal of unsupervised learning is to restructure the
input data into new features or a group of objects with
similar patterns. 41
 Unsupervised Learning…Cont”D
In unsupervised learning, we don't have a
predetermined result.
The
machine tries to find useful insights from the huge
amount of data.
Itcan be further classified into two categories of
algorithms: Definition: Clustering groups similar data points into clusters, without knowing the
categories in advance.

Clustering Goal: To identify natural groupings in data.

Definition: Association finds rules or relationships between variables in large datasets.

Association Goal: To discover interesting if-then relationships (also called association rules).

42
3) Reinforcement Learning
Reinforcement Learning is a feedback-based learning
method, in which a learning agent gets a reward for each
right action and gets a penalty for each wrong action.
The agent learns automatically with these feedbacks
and improves its performance.
Inreinforcement learning, the agent interacts with the
environment and explores it.

43
3) Reinforcement Learning…Cont”D
Thegoal of an agent is to get the most reward points,
and hence, it improves its performance.
The robotic dog, which automatically learns the
movement of his arms, is an example of Reinforcement
learning.

44
Applications of Machine Learning
Machine learning is a buzzword for today's technology,
and it is growing very rapidly day by day.
We are using machine learning in our daily life even
without knowing it such as Google Maps, Google
assistant, Alexa, etc.
The following are some of the most trending real-world
applications of Machine Learning.

45
Applications of Machine Learning

46
1. Image Recognition:
Image recognition is one of the most common
applications of machine learning. It is used to identify
objects, persons, places, digital images, etc.
The popular use case of image recognition and face
detection is Automatic friend tagging suggestion:
Facebook provides us a feature of auto friend tagging
suggestion.

47
 Image Recognition…Cont”D
Whenever we upload a photo with our Facebook
friends, then we automatically get a tagging suggestion
with name, and the technology behind this is machine
learning's face detection and recognition algorithm.
Itis based on the Facebook project named "Deep
Face," which is responsible for face recognition and
person identification in the picture.

48
2. Speech Recognition
While using Google, we get an option of "Search by voice,"
it comes under speech recognition, and it's a popular
application of machine learning.
Speech recognition is a process of converting voice
instructions into text, and it is also known as "Speech to
text", or "Computer speech recognition."
At present, machine learning algorithms are widely used by
various applications of speech recognition. Google
Assistant, Siri, Cortana, and Alexa are using speech
recognition technology to follow the voice instructions.
49
3. Traffic prediction:
Ifwe want to visit a new place, we take help of Google
Maps, which shows us the correct path with the shortest
route and predicts the traffic conditions.
Itpredicts the traffic conditions such as whether traffic is
cleared, slow-moving, or heavily congested with the help of
two ways:
 Real Time location of the vehicle form Google Map app
and sensors
 Average time has taken on past days at the same time.
50
 Traffic prediction…Cont”D
Everyone who is using Google Map is helping this app to
make it better.
Ittakes information from the user and sends back to its
database to improve the performance.

51
4. Product recommendations:
Machine learning is widely used by various e-commerce
and entertainment companies such as Amazon, Netflix,
etc., for product recommendation to the user.
Whenever we search for some product on Amazon, then we
started getting an advertisement for the same product while
Internet surfing on the same browser and this is because of
machine learning.

52
 Product recommendations…Cont”D
Google understands the user interest using various machine
learning algorithms and suggests the product as per
customer interest.
Similarly, when we use Netflix, we find some
recommendations for entertainment series, movies, etc., and
this is also done with the help of machine learning.

53
5. Self-driving cars:
One of the most exciting applications of machine learning
is self-driving cars.
Machine learning plays a significant role in self-driving
cars. Tesla, the most popular car manufacturing company is
working on self-driving car.
It
is using unsupervised learning method to train the car
models to detect people and objects while driving.

54
6. Email Spam and Malware Filtering:
Whenever we receive a new email, it is filtered
automatically as important, normal, and spam.
We always receive an important mail in our inbox with the
important symbol and spam emails in our spam box, and
the technology behind this is Machine learning. Below are
some spam filters used by Gmail:
Content Filter
Header filter
General blacklists filter
Rules-based filters
55
Permission filters
Email Spam and Malware Filtering…Cont”D
Some machine learning algorithms such as Multi-Layer
Perceptron, Decision tree, and Naïve Bayes classifier are
used for email spam filtering and malware detection.

56
7. Virtual Personal Assistant:
We have various virtual personal assistants such as Google
assistant, Alexa, Cortana, Siri. As the name suggests, they
help us in finding the information using our voice
instruction.
These assistants can help us in various ways just by our
voice instructions such as Play Music, call someone, Open
an email, Scheduling an appointment, etc.
Thesevirtual assistants use machine learning algorithms as
an important part.
57
 Machine learning Life cycle
 How does a machine learning system work?
 It can be described using the life cycle of machine learning

 Machine learning life cycle is a cyclic process to build an efficient

machine learning project.
 The main purpose of the life cycle is to find a solution to the problem
or project.
 Machine learning life cycle involves about seven major steps.
Gathering Data
Data preparation
Data Wrangling
Analyze Data
Train the model
Test the model
58
Deployment
ML Life cycle …Cont’D

59
 ML Life cycle …Cont’D
 The most important thing in the complete process is to
understand the problem and to know the purpose of the problem.
 In the complete life cycle process to solve a problem, we create a
machine learning system called "model", and this model is
created by providing "training".
 But to train a model, we need data. Hence, life cycle starts by
collecting data
1. Gathering Data:
Data gathering is the first step of the machine learning life cycle.
The goal of this step is to identify and obtain all data-related to the
60
problem.
 ML Life cycle …Cont’D
Gathering …
 Identify the sources.
 This step includes the following tasks:
 Identify various data sources
 Collect data
 Integrate the data obtained from different sources

61
 ML Life cycle …Cont’D
2. Data Preparation
 After collecting the data, we need to prepare it for further steps. Data
preparation is a step where we put our data into a suitable place and
prepare it to use in our machine learning training.
 In this step, first, we put all data together, and then randomize the
ordering of data.
 This step can be further divided into two processes:

 Data exploration: It is used to understand the nature of data that

we have to work with. A better understanding of data leads to an
effective outcome. In this, we find Correlations, general trends, and
outliers.
 Data pre-processing: Now the next step is preprocessing of data
for its analysis. 62
 ML Life cycle …Cont’D
3. Data Wrangling
Data wrangling is the process of cleaning and converting raw data
into a useable format.
In real-world applications, collected data may have various issues,
including:
Missing Values
Duplicate data
Invalid data
Noise
 So, we use various filtering techniques to clean the data. It is
mandatory to detect and remove the above issues because it can
negatively affect the quality of the outcome.
63
 ML Life cycle …Cont’D
[Link] Analysis
 This step involves:
Selection of analytical techniques
Building models
Review the result
The aim of this step is to build a machine learning model to
analyze the data using various analytical techniques.
 It starts with the determination of the type of the problems,. then
build the model using prepared data, and evaluate the model.
Hence, in this step, we take the data and use machine learning
algorithms to build the model. 64
 ML Life cycle …Cont’D
5. Training the Model
In this step we train our model to improve its performance for better
outcome of the problem.
We use datasets to train the model using various machine learning
algorithms.
Training a model is required so that it can understand the various
patterns, rules, and, features of the data.
6. Testing the Model
In this step, we check for the accuracy of our model by providing a
test dataset to it.
Testing the model determines the percentage accuracy of the
model as per the requirement of project or problem.
65
 ML Life cycle …Cont’D
7. Deployment
The last step of machine learning life cycle is deployment, where
we deploy the model in the real-world system.
Ifthe above-prepared model is producing an accurate result as per
our requirement with acceptable speed, then we deploy the model
in the real system.
But, before deploying the project, we will check whether it is
improving its performance using available data or not. The
deployment phase is similar to making the final report for a
project.
66
Difference between Data Science and Machine Learning

 Data Science and Machine Learning are the two popular modern
technologies, and they are growing with an immoderate rate.
 But these two buzzwords, along with artificial intelligence and deep
learning are very confusing term, so it is important to understand how
they are different from each other.
 Data Science and Machine Learning are closely related to each other
but have different functionalities and different goals.
 At a glance, Data Science is a field to study the approaches to find
insights from the raw data whereas,
 Machine Learning is a technique used by the group of data scientists to
enable the machines to learn automatically from the past data.
67
 Difference between Data Science and ML…Cont’D
To understand the difference in-depth, let's first have a brief
introduction to these two technologies.

68
 Skills required to be become a Data Scientist

 An excellent programming knowledge of Python, R, SAS, or Scala.

 Experience in SQL database Coding.
 Knowledge of Machine Learning Algorithms.
 Deep Knowledge of Statistics concepts.
 Data Mining, cleaning, and visualizing skills.
 Skills to use Big data tools such as Hadoop.

69
 Skills Needed for the Machine Learning Engineer:

 Understanding and implementation of Machine Learning Algorithms.

 Natural Language Processing.
 Good Programming knowledge of Python or R.
 Knowledge of Statistics and probability concepts.
 Knowledge of data modeling and data evaluation.

70
 What is Deep Learning?
Deep learning is defined as the subset of machine learning and
artificial intelligence that is based on artificial neural networks.
Indeep learning, the deep word refers to the number of layers in
a neural network.
Deep Learning is a set of algorithms inspired by the structure and
function of the human brain.
It uses a huge amount of structured as well as unstructured data
to teach computers and predicts accurate results.

71