Module 1

Introduction to Machine
Learning
Overview
⚫ Machine Learning
⚫ Types of Machine Learning
⚫ Issues in Machine Learning
⚫ Application of Machine Learning
⚫ Steps in developing a Machine Learning
Application.
What is Machine Learning ?
⚫ Machine learning is an area in computer
science which involves teaching
computers to do things naturally by
learning through their experience
⚫ It means training the computer to convert
data into information
⚫ “Learning happens through experience”-
ML algorithms adaptively increase their
performance
What is Machine Learning ?
⚫ Tom Mitchell provides a more
modern definition: “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.” ... To
find that logic is called “machine
learning
What is Machine Learning ?
⚫ A well defined learning problem requires:
◦ Class of task
◦ Performance measurement that needs to
be improved
◦ Source of experience
⚫ Example: A computer program that learns to
play tic-tac-toe ( or what we call X’s and 0’s)
◦ T: playing the game( i.e. making one row
column or diagonal X’s or 0’s)
◦ P: ability to win the game
◦ E: playing the game against itself
What is Machine Learning ?
⚫ Anotherexample: Robot
Navigation in a maize
◦T:
◦P:
◦E:
What is Machine Learning ?
⚫ Another example: Robot Navigation
in a maize
◦Navigating the maize from start to finish
by the robot
◦Reaching the end of the maize
◦Time taken to reach the maize
What is Machine Learning ?
⚫ Anotherexample: Robot
Navigation in a maize
◦T: Reaching the end of the maize
◦P: Time taken to reach the maize
◦E: Navigating the maize from start to
finish by the robot
What is Machine Learning ?
⚫ Yetanother example: Your email
program watches the way of
marking the emails as spam or not
◦T:
◦P:
◦E:
What is Machine Learning ?
⚫ Yet another example: Your email
program watches the way of marking
the emails as spam or not
◦T: Classifying the emails as spam or
not
◦P: the number of emails correctly
classified as spam/ or not spam
◦E: Watching you labeling the emails
as spam or not spam
Some common definitions:
• Nvidia : “Machine learning at its most basics is the
practice of using algorithms to parse data, learn
from it, and then make a determination or
prediction about something in the world”
• Stanford University: “Machine learning is the
science of getting computers to act without being
explicitly programmed.”
• McKinsey & Company states that “Machine
learning is based on algorithms that can learn
from data without relying on rules-based
programming”
Some common definitions:
• Pedro Domingos ( Computer Science
Department, University of Washington) in his
paper ‘A Few Useful Things to Know About
Machine Learning’ states “Machine learning
algorithms can figure out how to perform
important tasks by generalizing from
examples”
Some common definitions:
• According to the researchers in Carnegie Mellon
University’s ML department :The field of
Machine learning seeks to answer the question
‘ How can we build computer systems that
automatically improve with experience and
fundamental laws that govern all learning
processes?’
Components of ML study

Computer Statistics
Science

Engineering
Examples
⚫ Have you ever shopped
online? So while checking
for a product, did you
noticed when it
recommends for a product
similar to what you are
looking for? or did you
noticed “the person
bought this product also
bought this” combination
of products. How are they
doing this
recommendation? This is
machine learning.

https://www.edureka.co/blog/what-is-machine-learning/
Examples
⚫ Did you ever get a call
from any bank or finance
company asking you to
take a loan or an
insurance policy? What do
you think, do they call
everyone? No, they call
only a few selected
customers who they think
will purchase their
product. How do they
select? This is target
marketing and can be
applied using Clustering.
This is machine learning

https://www.edureka.co/blog/what-is-machine-learning/
How does ML algorithm work?
⚫ Machine learning uses algorithms to find
patterns in data and then uses a model
that recognizes those patterns to make
predictions on new data
How does ML algorithm work?
Types of Machine Learning
⚫ Machine learning is sub-categorized to three
types:
⚫ Supervised Learning – Train Me!
⚫ Unsupervised Learning – I am self sufficient
in learning
⚫ Reinforcement Learning – My life My rules!
(Hit & Trial)
⚫
What is Supervised Learning?
⚫ Supervised Learning is the one, where you
can consider the learning is guided by a
teacher.
⚫ We have a dataset which acts as a teacher
and its role is to train the model or the
machine.
⚫ Once the model gets trained it can start
making a prediction or decision when new
data is given to it.
⚫ Example: Classification of fruits
What is Supervised Learning?
What is Unsupervised Learning
⚫ The model learns through observation and
finds structures in the data.
⚫ Once the model is given a dataset, it
automatically finds patterns and
relationships in the dataset by creating
clusters in it.
⚫ What it cannot do is add labels to the
cluster, like it cannot say this a group of
apples or mangoes, but it will separate all
the apples from mangoes.
What is Unsupervised Learning
⚫ Suppose we presented images of apples,
bananas and mangoes to the model, so
what it does, based on some patterns and
relationships it creates clusters and
divides the dataset into those clusters.
⚫ Now if a new data is fed to the model, it
adds it to one of the created clusters.
What is Unsupervised Learning
What is Reinforcement Learning?
⚫ It is the ability of an agent to interact with the
environment and find out what is the best
outcome.
⚫ It follows the concept of hit and trial method.
⚫ The agent is rewarded or penalized with a point
for a correct or a wrong answer, and on the basis
of the positive reward points gained the model
trains itself.
⚫ And again once trained it gets ready to predict
the new data presented to it.
What is Reinforcement Learning?
Issues in Machine Learning
⚫ Having Algorithms Become Obsolete as
Soon as Data Grows. ...
⚫ Getting Bad Predictions to Come Together
With Biases. ...
⚫ Making the Wrong Assumptions- missing
data points
⚫ Receiving Bad Recommendations. ...
⚫ Having Bad Data Convert to Bad Results.
Application of Machine Learning
⚫ Virtual Personal Assistants
⚫ Predictions while Commuting
⚫ Videos Surveillance
⚫ Social Media Services
⚫ Email Spam and Malware Filtering
⚫ Online Customer Support
⚫ Search Engine Result Refining
⚫ Product Recommendations
⚫ Online Fraud Detection
Steps in developing a Machine
Learning Application
⚫ Step 1: Collect data
◦ collect sample by scraping a website and
extracting data
◦ Get information from RSS feed or API
⚫ Step 2: Prepare the input data
◦ Arrange the data in a standard format
(algorithm specific format)
◦ Example: string, integer
⚫ Step 3: Analyze the input data
◦ Recognize pattern
◦ Plotting the data in one or more dimension
depending on the number of features
Steps in developing a Machine
Learning Application
⚫ Step 4: Validating the data
◦ Ensure that there is no garbage data
◦ This step may be skipped if the we trust the
data source of if it is known how the data looks
like
⚫ Step 5: Train the algorithm
◦ Machine learning phase
◦ Feed good clean data to the algorithm and
extract knowledge or information
◦ For unsupervised learning there is no training
step
Steps in developing a Machine
Learning Application
⚫ Step 6: Test the algorithm
◦ Evaluating the algorithm how well it works
◦ Evaluate the success. If it is not satisfactory go
back to step 4
◦ Often there may be problem with collection of data,
then it is required to goto step 1
⚫ Step 7: Use it!
◦ Make a real program to do some task and see if all
previous steps work as expected
◦ If new data is encountered revisit step 1-5