© Jitesh Khurkhuriya – Azure ML Online Course

Basics of Machine Learning

© Jitesh Khurkhuriya – Azure ML Online Course

Why Machine Learning is the
Future?

© Jitesh Khurkhuriya – Azure ML Online Course

 Data Growth

Exabytes
45000
40,900
40000

35000

30000

25000

20000

15000

10000
7900
5000
130 1200
0
2005 2010 2015 2020
Exabytes

© Jitesh Khurkhuriya – Azure ML Online Course IDC Data Growth Forecast

 Data Growth

90%

90% of today’s data has been created in last two years alone

© Jitesh Khurkhuriya – Azure ML Online Course

 Benefits of Machine Learning

• Faster decisions

• Develop insights that are beyond human capabilities

• Act at the right time and take advantage of opportunities, converting

them into closed deals.

© Jitesh Khurkhuriya – Azure ML Online Course

Why Azure ML?

© Jitesh Khurkhuriya – Azure ML Online Course

 Why Azure ML?
• Drag and Drop interface and no Programming
required

• Large variety of algorithm as modules

• From experiment to production API in minutes

• Supports R and Python to bring in your existing

code

• Flexibility of data storage; supports variety of

data storage options

• Large number of pre-built APIs available as a

service

© Jitesh Khurkhuriya – Azure ML Online Course

What is Machine Learning?

© Jitesh Khurkhuriya – Azure ML Online Course

 What Is Machine Learning?
• Machine learning is the subfield of computer science that
gives computers the ability to learn without being explicitly
programmed.
– Arthur Samuel, 1959

• Extraction of knowledge from data

• Learns from past behaviour and make predictions or decisions

© Jitesh Khurkhuriya – Azure ML Online Course

 How Machines Learn?

Green Apple

Historic Data Learning Algorithms Intelligent Model

© Jitesh Khurkhuriya – Azure ML Online Course

 Supervised, Unsupervised
and Reinforcement Learning

© Jitesh Khurkhuriya – Azure ML Online Course

 Supervised Machine Learning
• Data is labelled
• There is an Input variable “X” or set of input variables and an output variable “Y”

Y = f(X)
• The function is approximated to predict new values of Y given X
• Examples
• Regression – Output variable is a real value such as Amount, Height, Weight
etc
• Classification – Output variable is a category, such as Yes, No, Red, Blue,
Yellow etc
Loan_ID Gender Married Dependents Self_Employed Income LoanAmt Term CreditHistory Property_Area Status
LP001002 Male No 0 No $5,849.00 60 1 Urban Y
LP001003 Male Yes 1 No $4,583.00 $128.00 120 1 Rural N
LP001005 Male Yes 0 Yes $3,000.00 $66.00 60 1 Urban Y
LP001006 Male Yes 2 No $2,583.00 $120.00 60 1 Urban Y

© Jitesh Khurkhuriya – Azure ML Online Course

 Unsupervised Machine Learning
• Only X or input variable is known

• The goal for unsupervised learning is to model the underlying

structure or distribution in the data in order to learn more
about the data.

• There is no correct answers and there is no teacher.

• Algorithms are left on their own to discover and present the

interesting structure in the data.

• Examples
• Clustering – Customer behaviour grouping
• Association – Recommendation model

© Jitesh Khurkhuriya – Azure ML Online Course

 Reinforcement Learning

• Reinforcement learning rewards good behaviour and penalizes bad

ones

• The idea is to maximise the gain or reward

Ad1 Ad2 Ad3 ……………. Ad N

© Jitesh Khurkhuriya – Azure ML Online Course

Understanding Data,
Variables/Features

© Jitesh Khurkhuriya – Azure ML Online Course

 Understanding The Variables Using a Dataset
Loan_ID Gender Married Dependents Self_Employed Income LoanAmt Term CreditHistory Property_Area Status
LP001002 Male No 0 No $5,849.00 60 1 Urban Y
LP001003 Male Yes 1 No $4,583.00 $128.00 120 1 Rural N
LP001005 Male Yes 0 Yes $3,000.00 $66.00 60 1 Urban Y
LP001006 Male Yes 2 No $2,583.00 $120.00 60 1 Urban Y

Types of Variables Data Type Category

• Predictor/Independent • Character/String • Categorical
• Gender • Gender • Gender
• Married • Married • Married
• Dependents • Self_Employed • Self-Employed
• Self_Employed • Property_Area • CreditHistory
• Income • Status • Property_Area
• LoanAmt • Status
• Term • Numeric
• CreditHistory • Dependents • Continuous
• PropertyArea • Income • Dependents
• LoanAmt • Income
• Target/Dependent • Term • LoanAmt
• Status • CreditHistory • Term

© Jitesh Khurkhuriya – Azure ML Online Course

Recap of Common Terms

© Jitesh Khurkhuriya – Azure ML Online Course

 Mean and Median
• Mean – Average of all the values
Mean = Sum of Salary/Number of observations
= 62,800/11
= $ 5709.09

• Median – Numerical Middle value of the sorted observations with

equal number of observations on both sides,

4,000 4,400 5,000 5,500 5,700 5,800 6,200 6,400 6,400 6,400 7,000

1 2 3 4 5 6 7 8 9 10 11

© Jitesh Khurkhuriya – Azure ML Online Course

 Mode and Range
• Mode - The value that appears most often in a set of data
6,400

• Range – The difference of highest and lowest values in a

sample of observations
7000 – 4000 = 3,000

© Jitesh Khurkhuriya – Azure ML Online Course

 Probability
• Probability is a numerical way of describing how likely something is
going to happen.

• Sample Space (S) – Set of possible outcomes that might be observed for
an event
• Dice Sample Space (S) = {1, 2, 3, 4, 5, 6}
• Probability of 3
• P(A) = 1/6 = 0.1667

• Probability of getting an even number from the given sample space

• How many even numbers are there? 2,4,6
• So number of even occurrences = 3
• Probability of getting an even number is P(A) = 3/6 = 0.5 or 50%

© Jitesh Khurkhuriya – Azure ML Online Course

Types of Models

© Jitesh Khurkhuriya – Azure ML Online Course

 Classification
• Identification of category of data Y

• Binary/Two-Class Classification – Either/Or, Yes or No type

• Multi-Class Classification – One of the many alternatives

• Examples
• Assigning a given email into "spam" or "non-
spam" classes Or Primary, Social or Promotional
emails
• Will this customer default on loan repayment?
X
• Will this customer buy my product?

Predicting the value for categorical variable.

© Jitesh Khurkhuriya – Azure ML Online Course

 Regression Analysis
Regression Line
• Estimating the relationships among variables Y Y =f(X)
Predicted Value

• Predictor is a continuous variable

Claims
• Examples
• Predicting the future sale of products
• Computing fair price of the product or
service

• One of the most common methods used in Number of Medical

Machine Learning Claims
Age X

• Infer causal relationships between dependent

and independent variables.

© Jitesh Khurkhuriya – Azure ML Online Course

 Clustering or Cluster Analysis
• Clustering is the task of grouping a set of objects in Y
such a way that

• objects in the same group (called a cluster) are

• more similar (in some sense or another)
• to each other than
• to those in other groups (clusters)

• Unsupervised Learning model

• Customers who make lot of long-distance calls and

don’t have a job. Who are they? X

© Jitesh Khurkhuriya – Azure ML Online Course

 Anomaly Detection
• Anomaly detection (also outlier detection) is the Y
• Identification of items, events or observations which
• Do not conform to an expected pattern or other
items in a dataset.

• Typically the anomalous items will translate to some

kind of problem such as
• Bank fraud
• Credit Card Fraud
• Structural defect
• Medical problems
X
• Anomalies are also referred to as outliers, novelties,
noise, deviations and exceptions.

© Jitesh Khurkhuriya – Azure ML Online Course

Thank You and Have a Great Time!

© Jitesh Khurkhuriya – Azure ML Online Course