UNIVERSITY INSTITUTE OF

TECHNOLOGY
BARKATULLAH UNIVERSITY BHOPAL

MACHINE LEARNIG
SESSION 2017-18

SUBMITTED TO: SUBMITTED BY:

KAMINI MAAM ADARSH PATIDAR
ACKNOWLEDGMENT
I have taken efforts in this project however it would not
have been possible without the kind support and help of
many individuals. I would like to extend my sincere
thanks to all of them.
I am highly indebted to UNIVERSITY INSTITUE OF
TECHNOLOGY BARKATULLAH UNIVERSITY BHOPAL, for
their guidance and constant supervision as well as for
providing necessary information regarding the project
and also for their support in completing the project.
Special thanks to Ms. Kamini Maheswari and Mrs.
Vasima Khan.
I would like to express my gratitude towards my parents
and members of BUIT for their kind cooperation and
encouragement which helped me in completion of this
project.
My thanks and appreciation also goes to my colleagues in
developing this project and people who have willingly
helped me out with their abilities.
ADARSH PATIDAR
 (CS IIND SEM)
CONTENTS:

MACHINE LEARNING

HISTORY OF MACHINE LEARNING

EVOLUTION OF MACHINE LEARNING

IMPORTANCE OF MACHINE LEARNING

METHODS OF MACHINLE LEARNING

WORKING

NEUTRAL NETWORKS AND MACHINE LEARNING

DIFFERENCE WITH RELATED FIELDS

APPLICATIONS

CONCLUSION

BIBLIOGRAPHY
 Machine Learning

Definition - Machine learning is a method of data analysis that automates analytical

model building. Using algorithms that iteratively learn from data, machine learning
allows computers to find hidden insights without being explicitly programmed where to
look.

Computers have helped us to calculate the vastness of space and the minute details of
subatomic particles. When it comes to counting and calculating, or following logical
yes/no algorithms computers outperform humans thanks to the electrons moving
through their circuitry at the speed of light. But we generally dont consider them as
intelligent because, traditionally, computers havent been able to do anything
themselves, without being taught (programmed) by us first.

So far, even if a computer had access to all of the information in the world it couldnt do
anything smart with it. It could find us a picture of a cat but only because we had told
it that certain pictures contain cats. In other words, ask it to find a picture of a cat and it
will return with a picture which it has been told is of a cat.

This has several implications which limit its helpfulness not least that a large amount of
human time has to be spent telling it what every picture contains. The data (pictures) need
to pass through a human bottleneck, where they are labeled, before the computer can,
with lightning-quick precision, identify it as a cat picture and show it to us when we
request it.

While this works well enough if we are just searching for cat pictures on Google to pass
our time, if we want to do something more advanced such as monitor a live video feed
and tell us when a cat wanders in front of the camera its not so great.

It is problems like this which machine learning is trying to solve. At its most simple,
machine learning is about teaching computers to learn in the same way we do, by
interpreting data from the world around us, classifying it and learning from its successes
and failures. In fact, machine learning is a subset, or better, the leading edge of artificial
intelligence.

Machine learning is the subfield of computer science that, according to Arthur Samuel
in 1959, gives "computers the ability to learn without being explicitly programmed."
Evolved from the study of pattern recognition and computational learning theory in
artificial intelligence, machine learning explores the study and construction of algorithms
that can learn from and make predictions on data such algorithms overcome following
strictly static program instructions by making data-driven predictions or decisions,
through building a model from sample inputs. Machine learning is employed in a range
of computing tasks where designing and programming explicit algorithms with good
performance is difficult or infeasible; example applications include email filtering,
detection of network intruders or malicious insiders working towards a data breach,
optical character recognition (OCR), learning to rank, and computer vision.

Machine learning is closely related to (and often overlaps with) computational statistics,
which also focuses on prediction-making through the use of computers. It has strong ties
to mathematical optimization, which delivers methods, theory and application domains to
the field. Machine learning is sometimes conflated with data mining, where the latter
subfield focuses more on exploratory data analysis and is known as unsupervised
learning. Machine learning can also be unsupervised and be used to learn and establish
baseline behavioral profiles for various entities and then used to find meaningful
anomalies.

Within the field of data analytics, machine learning is a method used to devise complex
models and algorithms that lend themselves to prediction; in commercial use, this is
known as predictive analytics. These analytical models allow researchers, data scientists,
engineers, and analysts to "produce reliable, repeatable decisions and results" and
uncover "hidden insights" through learning from historical relationships and trends in the
data.

As of 2016, machine learning is a buzzword, and according to the Gartner hype cycle of
2016, at its peak of inflated expectations. Effective machine learning is difficult because
finding patterns is hard and often not enough training data is available; as a result,
machine-learning programs often fail to deliver.

Tom M. Mitchell provided a widely quoted, more formal definition of the algorithms
studied in the Machine Learning field: "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." This
definition of the tasks in which machine learning is concerned offers a fundamentally
operational rather than defining the field in cognitive terms. This follows Alan Turing's
proposal in his paper "Computing Machinery and Intelligence", in which the question
"Can machines think?" is replaced with the question "Can machines do what we (as
thinking entities) can do.

History of Machine Learning

Timeline of Machine Learning

1950 Alan Turing creates the Turing Test to determine if a computer has real
intelligence. To pass the test, a computer must be able to fool a human into believing it is
also human.

1952 Arthur Samuel wrote the first computer learning program. The program was the
game of checkers, and the IBM IBM computer improved at the game the more it played,
studying which moves made up winning strategies and incorporating those moves into its
program.

1957 Frank Rosenblatt designed the first neural network for computers (the
perceptron), which simulate the thought processes of the human brain.

1967 The nearest neighbor algorithm was written, allowing computers to begin
using very basic pattern recognition. This could be used to map a route for traveling
salesmen, starting at a random city but ensuring they visit all cities during a short tour.

1979 Students at Stanford University invent the Stanford Cart which can navigate
obstacles in a room on its own.

1981 Gerald Dejong introduces the concept of Explanation Based Learning (EBL), in
which a computer analyses training data and creates a general rule it can follow by
discarding unimportant data.

1985 Terry Sejnowski invents NetTalk, which learns to pronounce words the same
way a baby does.
1990s Work on machine learning shifts from a knowledge-driven approach to a data-
driven approach. Scientists begin creating programs for computers to analyze large
amounts of data and draw conclusions or learn from the results.

1997 IBMs Deep Blue beats the world champion at chess.

2006 Geoffrey Hinton coins the term deep learning to explain new algorithms that
let computers see and distinguish objects and text in images and videos.

2010 The Microsoft MSFT Kinect can track 20 human features at a rate of 30 times
per second, allowing people to interact with the computer via movements and gestures.

2011 IBMs Watson beats its human competitors at Jeopardy.

2011 Google Brain is developed, and its deep neural network can learn to discover and
categorize objects much the way a cat does.

2012 Googles X Lab develops a machine learning algorithm that is able to

autonomously browse YouTube videos to identify the videos that contain cats.

2014 Facebook develops DeepFace, a software algorithm that is able to recognize or

verify individuals on photos to the same level as humans can.

2015 Amazon launches its own machine learning platform.

2015 Microsoft creates the Distributed Machine Learning Toolkit, which enables the
efficient distribution of machine learning problems across multiple computers.

2015 Over 3,000 AI and Robotics researchers, endorsed by Stephen Hawking, Elon
Musk and Steve Wozniak (among many others), sign an open letter warning of the
danger of autonomous weapons which select and engage targets without human
intervention.

2016 Googles artificial intelligence algorithm beats a professional player at the

Chinese board game Go, which is considered the worlds most complex board game and
is many times harder than chess. The AlphaGo algorithm developed by Google
DeepMind managed to win five games out of five in the Go competition.

So are we drawing closer to artificial intelligence? Some scientists believe thats actually
the wrong question.

They believe a computer will never think in the way that a human brain does, and that
comparing the computational analysis and algorithms of a computer to the machinations
of the human mind is like comparing apples and oranges.
 Evolution of machine learning
Because of new computing technologies, machine learning today is not like machine
learning of the past. It was born from pattern recognition and the theory that computers
can learn without being programmed to perform specific tasks; researchers interested in
artificial intelligence wanted to see if computers could learn from data. The iterative
aspect of machine learning is important because as models are exposed to new data, they
are able to independently adapt. They learn from previous computations to produce
reliable, repeatable decisions and results. Its a science thats not new but one thats
gaining fresh momentum.

While many machine learning algorithms have been around for a long time, the ability to
automatically apply complex mathematical calculations to big data over and over, faster
and faster is a recent development. Here are a few widely publicized examples of
machine learning applications you may be familiar with:

The heavily hyped, self-driving Google car? The essence of machine learning.
Online recommendation offers such as those from Amazon and Netflix? Machine
learning applications for everyday life.
Knowing what customers are saying about you on Twitter? Machine learning
combined with linguistic rule creation.
Fraud detection? One of the more obvious, important uses in our world today.

How did machine learning come about?

Building algorithms capable of doing this, using the binary yes and no logic of
computers, is the foundation of machine learning a phrase which was probably first
used during serious research by Arthur Samuel at IBM during the 1950s. Samuels
earliest experiments involved teaching machines to learn to play checkers.

As knowledge something to draw insight from and a basis for making decisions is
deeply integral to learning, these early computers were severely handicapped due to the
lack of data at their disposal. Without all of the digital technology we have today to
capture and store information from the analogue world, machines could only learn from
data slowly inputted through punch cards and, later, magnetic tapes and storage.
Today things are a little different thanks to the rollout of the internet, the proliferation
of mobile, data-gathering phones and other devices and the adoption of online, connected
technology in industry, we literally have more data than we know how to deal with.

No human brain can hope to process even a fraction of the digital information it has
available. But, with its lightning speed and infallible binary logic, could a computer.

Why is machine learning important?

Resurging interest in machine learning is due to the same factors that have made data
mining and Bayesian analysis more popular than ever. Things like growing volumes and
varieties of available data, computational processing that is cheaper and more powerful,
and affordable data storage.

All of these things mean it's possible to quickly and automatically produce models that
can analyze bigger, more complex data and deliver faster, more accurate results even on
a very large scale. And by building precise models, an organization has a better chance of
identifying profitable opportunities or avoiding unknown risks.

Machine learning has several very practical applications that drive the kind of real
business results such as time and money savings that have the potential to
dramatically impact the future of your organization. At Interactions in particular, we see
tremendous impact occurring within the customer care industry, whereby machine
learning is allowing people to get things done more quickly and efficiently. Through
Virtual Assistant solutions, machine learning automates tasks that would otherwise need
to be performed by a live agent such as changing a password or checking an account
balance. This frees up valuable agent time that can be used to focus on the kind of
customer care that humans perform best: high touch, complicated decision-making that is
not as easily handled by a machine. At Interactions, we further improve the process by
eliminating the decision of whether a request should be sent to a human or a machine:
unique Adaptive Understanding technology, the machine learns to be aware of its
limitations, and bail out to humans when it has a low confidence in providing the correct
solution.

Machine learning has made dramatic improvements in the past few years, but we are still
very far from reaching human performance. Many times, the machine needs the
assistance of human to complete its task. At Interactions, we have deployed Virtual
Assistant solutions that seamlessly blend artificial with true human intelligence to deliver
the highest level of accuracy and understanding.
 What are some popular machine learning
methods?

Supervised learning algorithms are trained using labeled examples, such as an input
where the desired output is known. For example, a piece of equipment could have data
points labeled either F (failed) or R (runs). The learning algorithm receives a set of
inputs along with the corresponding correct outputs, and the algorithm learns by
comparing its actual output with correct outputs to find errors. It then modifies the model
accordingly. Through methods like classification, regression, prediction and gradient
boosting, supervised learning uses patterns to predict the values of the label on additional
unlabeled data. Supervised learning is commonly used in applications where historical
data predicts likely future events. For example, it can anticipate when credit card
transactions are likely to be fraudulent or which insurance customer is likely to file a
claim.

Unsupervised learning is used against data that has no historical labels. The system is
not told the "right answer." The algorithm must figure out what is being shown. The goal
is to explore the data and find some structure within. Unsupervised learning works well
on transactional data. For example, it can identify segments of customers with similar
attributes who can then be treated similarly in marketing campaigns. Or it can find the
main attributes that separate customer segments from each other. Popular techniques
include self-organizing maps, nearest-neighbor mapping, k-means clustering and singular
value decomposition. These algorithms are also used to segment text topics, recommend
items and identify data outliers.

Semisupervised learning is used for the same applications as supervised learning. But it
uses both labeled and unlabeled data for training typically a small amount of labeled
data with a large amount of unlabeled data (because unlabeled data is less expensive and
takes less effort to acquire). This type of learning can be used with methods such as
classification, regression and prediction. Semisupervised learning is useful when the cost
associated with labeling is too high to allow for a fully labeled training process. Early
examples of this include identifying a person's face on a web cam.

Reinforcement learning is often used for robotics, gaming and navigation. With
reinforcement learning, the algorithm discovers through trial and error which actions
yield the greatest rewards. This type of learning has three primary components: the agent
(the learner or decision maker), the environment (everything the agent interacts with) and
actions (what the agent can do). The objective is for the agent to choose actions that
maximize the expected reward over a given amount of time. The agent will reach the goal
much faster by following a good policy. So the goal in reinforcement learning is to learn
the best policy.

How it works?
To get the most value from machine learning, you have to know how to pair the best
algorithms with the right tools and processes.

A core objective of a learner is to generalize from its experience. Generalization in this

context is the ability of a learning machine to perform accurately on new, unseen
examples/tasks after having experienced a learning data set. The training examples come
from some generally unknown probability distribution (considered representative of the
space of occurrences) and the learner has to build a general model about this space that
enables it to produce sufficiently accurate predictions in new cases.

The computational analysis of machine learning algorithms and their performance is a

branch of theoretical computer science known as computational learning theory. Because
training sets are finite and the future is uncertain, learning theory usually does not yield
guarantees of the performance of algorithms. Instead, probabilistic bounds on the
performance are quite common. The biasvariance decomposition is one way to quantify
generalization error.

For the best performance in the context of generalization, the complexity of the
hypothesis should match the complexity of the function underlying the data. If the
hypothesis is less complex than the function, then the model has underfit the data. If the
complexity of the model is increased in response, then the training error decreases. But if
the hypothesis is too complex, then the model is subject to overfitting and generalization
will be poorer.

In addition to performance bounds, computational learning theorists study the time

complexity and feasibility of learning. In computational learning theory, a computation is
considered feasible if it can be done in polynomial time. There are two kinds of time
complexity results. Positive results show that a certain class of functions can be learned
in polynomial time. Negative results show that certain classes cannot be learned in
polynomial time.
Neutral Networks and Machine Learning

No discussion of ML would be complete without at least mentioning neural networks.

Not only do neural nets offer an extremely powerful tool to solve very tough problems,
but they also offer fascinating hints at the workings of our own brains, and intriguing
possibilities for one day creating truly intelligent machines.

Neural networks are well suited to machine learning problems where the number of
inputs is gigantic. The computational cost of handling such a problem is just too
overwhelming for the types of systems weve discussed above. As it turns out, however,
neural networks can be effectively tuned using techniques that are strikingly similar to
gradient descent in principle.

What are the differences between data

mining, machine learning and deep learning?
Although all of these methods have the same goal to extract insights, patterns and
relationships that can be used to make decisions they have different approaches and
abilities.

Data Mining

Data mining can be considered a superset of many different methods to extract

insights from data. It might involve traditional statistical methods and machine
learning. Data mining applies methods from many different areas to identify
previously unknown patterns from data. This can include statistical algorithms,
machine learning, text analytics, time series analysis and other areas of analytics.
Data mining also includes the study and practice of data storage and data
manipulation.

Machine Learning

The main difference with machine learning is that just like statistical models, the
goal is to understand the structure of the data fit theoretical distributions to the
data that are well understood. So, with statistical models there is a theory behind
the model that is mathematically proven, but this requires that data meets certain
strong assumptions too. Machine learning has developed based on the ability to
use computers to probe the data for structure, even if we do not have a theory of
 what that structure looks like. The test for a machine learning model is a validation
error on new data, not a theoretical test that proves a null hypothesis. Because
machine learning often uses an iterative approach to learn from data, the learning
can be easily automated. Passes are run through the data until a robust pattern is
found.

Deep learning

Deep learning combines advances in computing power and special types of neural
networks to learn complicated patterns in large amounts of data. Deep learning
techniques are currently state of the art for identifying objects in images and words
in sounds. Researchers are now looking to apply these successes in pattern
recognition to more complex tasks such as automatic language translation, medical
diagnoses and numerous other important social and business problems.

What can be done with machine learning?

The application of machine learning to society and industry is leading to advancements
across many fields of human endeavour. For example, in medicine, machine learning is
being applied to genomic data to help doctors understand, and predict, how cancer
spreads, meaning more effective treatments can be developed.

Data from deep space is being collected here on Earth through huge radio telescopes
and after being analyzed with machine learning, is helping us to unlock the secrets of
black holes.

In retail, machine learning matches shoppers with products they want to buy online, and
in the bricks n mortar world it allows shop assistants to personalize the service they
offer their customers.

In the war against terror and extremism, machine learning is used to predict the behavior
of those wanting to harm the innocent.

In our day-to-day lives, machine learning now powers Googles search and image
algorithms, to more accurately match us with the information we need in our lives, at the
time we need it.

The process of allowing computers to understand and communicate with us in human

language, thanks to machine learning, is known as natural language processing (NLP)
and this has led to breakthroughs in translation technology and the voice controlled
devices we increasingly use every day, including Amazons Echo.
Without a doubt, machine learning is proving itself to be a technology with far-reaching
transformative powers. The science fiction dream of robots capable of working alongside
us and augmenting our own inventiveness and imagination with their flawless logic and
superhuman speed is no longer a dream it is becoming a reality in many fields. Machine
learning is the key which has unlocked it, and its potential future applications are almost
unlimited.

Some of the fields where it is presently used are:

1. Fighting Webspam

Its no secret that Google hates web spam. It damages the user experience and makes
them look bad by showing it as a relevant result. Googles ongoing war with spam is
well-documented, but evidence points to them using machine learning to amp up their
efforts.

To fight spam both online and in email, Google is using a neural network they call deep
learning. Instead of fighting spam through simple blacklists, this type of machine
learning uses data from users and from natural-language processing to draw conclusions
about the emails it looks at. Its not perfect, but the process is getting better by the day.

2. Imitation Learning

This type of machine learning is similar to observational learning, which is something

humans do as infants. It was posited that this type of learning could be used in humanoid
robots as far back as 1999.

Today, this is used in field robotics industries like construction, agriculture, search and
rescue, military, and others. In these situations, manually programming robotic solution is
much more difficult. Instead, collaborative methods like programming by
demonstration are used in conjunction with machine learning to teach programming in
the field.

3. Assistive and Medical Tech

Assistive robots, according to David L. Jaffe of Stanford, are devices that can sense,
process sensory information, and perform actions for people with disabilities and seniors,
along with drivers through the proper tools.

Through this type of machine learning, and real-world collaborations, the Smart Tissue
Autonomous Robot (STAR) was created. By using machine learning and 3D sensing, this
device has been able to stitch together pig intestines (used for testing) better than any
surgeon. While STAR isnt meant to replace surgeons, it does offer a collaborative
solution for delicate steps in various medical procedures.

Machine learning even has medical applications in the form of predictive measures.
Much in the same way that a colleague can look at a doctors patient notes and spot
things they may have missed, so too can an A.I look for patterns that point to possible
heart failure.

In this way, doctors can collaborate with the virtual robot A.I to better diagnose this fatal
heart condition before it strikes. The extra pair of eyes can save lives, and studies thus far
have shown promise for the future application of this technology.

4. Automatic Translation

While a simple concept, machine learning can also be used to instantly translate text into
another language. Not only this, but it can do the same thing with text on images! In the
case of text, the algorithm can learn about how words fit together and translate more
accurately.

In the case of images, the neural network identifies letters in the image, pulls them into
text, and then does the translation before putting them back into the picture.

Conclusion
Weve covered much of the basic theory underlying the field of Machine Learning here,
but of course, we have only barely scratched the surface.

Keep in mind that to really apply the theories contained in this introduction to real life
machine learning examples, a much deeper understanding of the topics discussed herein
is necessary. There are many subtleties and pitfalls in ML, and many ways to be lead
astray by what appears to be a perfectly well-tuned thinking machine. Almost every part
of the basic theory can be played with and altered endlessly, and the results are often
fascinating. Many grow into whole new fields of study that are better suited to particular
problems.

Clearly, Machine Learning is an incredibly powerful tool. In the coming years, it

promises to help solve some of our most pressing problems, as well as open up whole
new worlds of opportunity. The demand for ML engineers is only going to continue to
grow, offering incredible chances to be a part of something big. I hope you will consider
getting in on the action.
 BIBLIOGRAPHY
SEARCH ENGINES
WIKIPEDIA
GOOGLE
YAHOO
BING
BOOKS
Understanding Machine Learning: From Theory to
Algorithms by Shai Ben-David and Shai Shalev-
Shwartz