MACHINE LEARNING

Field of study
ABSTRACT
 Machine learning is the Scientific study of algorithms and statistical
models that computer
 Systems use to perform a specific task without using explicit
instructions . Machine learning was coined in 1959 by ARTHUR SAMUEL .
In my presentation u can see the complete overview about the machine
learning and u can see the different types of machine learning like
supervised machine learning ,unsupervised machine learning , active
machine learning and Reinforcement machine learning and example for
each type of it and different applications like Google maps, facial
recognition in smartphone and tagging of friend in Facebook and twitter
etc., which are in the usage , about machine learning projects like Google
Brain, Apache Mahout etc., and Comparison between artificial
intelligence and machine learning , that’s what machine learning is the
subset of the artificial intelligence.
Relation between ARTIFICAL INTELLIGENCE and MACHINE
LEARNING
Definition
 Machine learning is the scientific study of algorithms and
statistical models that computer systems use to perform a specific
task without using explicit instructions.

 simply we can say that we humans learn from the past

experiences and the machines follow the instructions given
by the humans but if humans can train the machines to
learn from the past data that’s what the machine learning
i.e., machines to learn from the given data
 introduction
 The name machine learning was coined in 1959 by ARTHUR SAMUEL.

 Here we uses DATA MINING in the machine learning . Data mining is

nothing but the practice of examining large pre-existing databases in
order to generate new information.

 we give large amount of data for machine to learn from the data but
useful information was taken by the machine using data mining . It plays a
major role in machine learning

 From the useful data as per the statistics we can predict the any kind of
situations

Example: when a company with huge amount of customers and they are
going to launch a new product if they give the data of the product and all
customers details . By using machine learning we can predict whether their
product will be success or not
 One of the best example/application to explain the MACHINE LEARNING

SPEECH RECOGNITION
 let us see the working of speech recognition
 First by using microphone we will give the audio/voice as a input

 Then it as per the amplitude and frequency check the words in the
database find the similar amplitude words in the database and display
the complete text

 Thus the speech converts in to the text and then browse and check for
information in the text

 If information is found then it displays on the screen

 This what we see in the above example/application and it is one of the

best application using machine learning

 There are many application related to this but I took this for
explanation of the working of the machine learning.
Difference between Traditional programming and Machine learning
DIFFERENT TYPES IN MACHINE
LEARNING
a) supervised learning

b) Unsupervised learning

c) Reinforcement learning
 SUPERVISED
 LEARNING
Supervised learning Is the machine learning task of learning a function
that maps an input to an output based on example of input-output pairs

 It infers a function from labeled training data consisting of a set of training

examples

 The data given in this labeled. Mathematically as y=f(x)

Example: If three different currency coins are given as a training data with
their weights as there feature and currency as a label . If u give a
coin to the machine of supervised learning then machine check its
feature and display its currency

 A Real time examples where we are using this supervised learning are
Fraud detection, Email spam detection , Diagnostics , Image Recognition,
Risk Assessment , score prediction.
 A clear idea about supervised learning u can see in the above
representation
 UNSUPERVISED
 LEARNING
Unsupervised learning is a type of machine learning
that looks for previously
undetected patterns in a data set with no pre-existing labels and with a
minimum of human supervision.

 In contrast to supervised learning that usually makes use of human-labeled

data

 Unsupervised learning also known as self-Organization

Example: If we give the data of different names of cricketers with their scores
and wickets then the machine of unsupervised learning with give the
separate data of batsman and bowlers based scores and wickets here
we are not given any information but it can analyze.

 A Real life time examples where we are using this unsupervised learning are
Text mining , Face Recognition , Big data visualization , image Recognition .
 A clear idea about unsupervised learning u can see in the above
representation
Difference between supervised and unsupervised learning
 REINFORCEMENT LEARNING
 Reinforcement learning is an area of machine learning and thereby also a
branch of Artificial intelligence. It allows machines and software agents to
automatically determine the ideal behavior within a specific context, in order
to maximize its performance.

 It works with the feedback when input is given and check for the output it
gives wrong output we give the feedback as wrong from that the machine can
start learn and gives the correct output

 A Best example already we seen in the real world is ALPHA GO has defeated
world champion KE JIE in the first game of a best –of-three match series in
china.
 Alpha go play itself as many times it receives feedback from itself and
defeated the world champion

Examples: Gaming , Finance , Manufacturing , Inventory Management ,

Robot Navigation.
 A clear idea about reinforcement learning is seen in the above
diagrammatic representation
 applications
Top 10 applications that are using machine learning in our daily life
activities
• Image Recognition
• Speech Recognition
• Medical Diagnosis
• Traffic Alerts
• Social Media
• Self Driving Cars
• Google Translate
• Fraud Detection
• Transportation and Commuting
• Online Video Streaming
• Products Recommendations
 we can seen many more applications in different sectors
Major applications of machine learning in our phones

Google Go/Ok Google/Google Assistant

Amazon Alexa

YouTube Recommendation system

Voice Recognition

Face Recognition
 FUTURE OF THIS MACHINE
LEARNING
The field of machine learning is experiencing exponential growth today ,
especially in the subject of computer vision. Today the error rate in humans is
only 3% in computer vision . This means computers are already better at
recognizing and analyzing images than humans.

The progress we have made from 26% error in 2011 to 3% error in 2016 is
hugely impactful. The way I like to think is, computers have now evolved eyes
that work.

Machine learning is the subset of Artificial Intelligence. If AI works along

with the Machine learning and Deep learning we can create a revolution of
ROBOTS in the history

Let see what will happen

We cannot say what will happen if machines
learn everything about the environment and
complete human life

MACHINES CAN MAKE THE PROBLEMS &

CAN SOLVE THE PROBLEMS

A Complete machine IQ (with AI , ML & DL )is

1000 times of a human being IQ
 THE NEXT LEVEL
ARTIFICIAL INTELLLIGENCE

THANK YOU

BY
SANKURU . NARMADA
III – year, E.C.E -C
UNIVERSAL COLLEGE OF ENGINEERING AND TECHNOLOGY