DEPARTMENT OF PALI & BUDDHIST STUDIES

UNIVERSITY OF DHAKA

REPORT ON MACHINE LEARNING

Md. Abir Hasan [12715 ]

[Reg.-2021816082]
2nd Year 4th Semester
Computer Science - 206

Dr. Ahmedul Kabir

Associate Professor
IIT, University of Dhaka
 Executive Summary

Here is a short description of what I’m going to discuss:

• Introduction to Computer Science, Human Intelligence & Artificial

Intelligence (AI).
• What is ML
• History of Machine Learning (ML)
• Types of ML
• Major Applications of ML
• The benefits of ML to our new Artificial Intelligence (AI) based
world.
 Computer Science, Human Intelligence & AI

Computer science focuses on the development and

testing of software and software systems. It involves
working with mathematical models, data analysis and
security, algorithms, and computational theory
This is the foundation of all computing disciplines. It
includes an entire range of subareas, like machine learning,
natural language processing, computing systems, networking, operating
systems, AI, and factors. human (website, Michigan Technological University)

We call ourselves Homo sapiens—man the wise—because

our intelligence is so important to us. For thousands of
years, we have tried to understand how we think; that is,
how a mere handful of matter can perceive, understand,
predict, and manipulate a world far larger and more
complicated than itself.
(Artificial Intelligence :A Modern Approach, Third Edition, Stuart Russell and Peter Norvig)

Artificial intelligence, or AI, is technology that

enables computers and machines to simulate
human intelligence and problem-solving
capabilities. As a field of computer science,
artificial intelligence encompasses (and is often
mentioned together with) machine
learning and deep learning. These disciplines involve the development of AI
algorithms, modeled after the decision-making processes of the human brain,
that can ‘learn’ from available data and make increasingly more accurate
classifications or predictions over time.
(Website, IBM)

Machine Learning

(This photos is downloaded from IBM website)

We are drowning in information and starving for knowledge. — John Naisbitt

In particular, we define machine learning as a set of methods that can

automatically detect patterns in data, and then use the uncovered patterns to
predict future data, or to perform other kinds of decision making under
uncertainty (such as planning how to collect more data!).
We are entering the era of big data. For example, there are about 1 trillion
web pages1; one hour of video is uploaded to YouTube every second,
amounting to 10 years of content every day2; the genomes of 1000s of
people, each of which has a length of 3.8 × 109 base pairs, have been
sequenced by various labs; Walmart handles more than 1M transactions per
hour and has databases containing more than 2.5 petabytes (2.5 × 1015) of
information (Cukier 2010); and so on.
This deluge of data calls for automated methods of data analysis, which is
what machine learning provides.
(Machine Learning : A Probabilistic Perspective, Kevin P. Murphy)

History of Machine Learning

Until the late 1970s, it was a part of AI’s evolution. Machine learning is, in
part, based on a model of brain cell interaction. The model was created in
1949 by Donald Hebb in a book titled “The Organization of Behavior.” The
book presents Hebb’s theories on neuron excitement and communication
between neurons.
The term machine learning was coined in 1959 by Arthur Samuel,
an IBM employee and pioneer in the field of computer
gaming and artificial intelligence. The synonym self-teaching
computers was also used in this time period.
Although the earliest machine learning model was introduced in the 1950s
when Arthur Samuel invented a program that
calculated the winning chance in checkers for
each side, the history of machine learning roots
back to decades of human desire and effort to
study human cognitive processes.
By the early 1960s an experimental "learning
machine" With punched tape memory, called Cybertron, had been
developed by Raytheon
Company toanalyze sonar signals,electrocardiograms, and speech patterns
using rudimentary reinforcement learning.
Interest related to pattern recognition continued into the 1970s. Tom M.
Mitchell provided a widely quoted, more formal definition of the algorithms
studied in the machine learning field.
In the late 1970s and early 1980s, artificial intelligence research focused on
using logical, knowledge-based approaches rather than algorithms.
Additionally, neural network research was abandoned by computer science
and AI researchers. This caused a schism between artificial intelligence and
machine learning.
(Source: Wikipedia)

Types of ML

Machine learning is usually divided into two main types.

1. Predictive or supervised learning approach.

2. Descriptive or unsupervised learning approach.
There is a third type of machine learning, known as reinforcement
learning, which is somewhat less commonly used.

1. Supervised Machine Learning

Supervised learning, also
known as supervised
machine learning, is defined
by its use of labeled datasets
to train algorithms to
classify data or predict
outcomes accurately. As
input data is fed into the
model, the model adjusts its
weights until it has been fitted appropriately. This occurs as part of
the cross validation process to ensure that the model
avoids overfitting or underfitting. Supervised learning helps
organizations solve a variety of real-world problems at scale, such
as classifying spam in a separate folder from your inbox. Some
methods used in supervised learning include neural networks, naïve
bayes, linear regression, logistic regression, random forest, and
support vector machine (SVM). There is a example of problem in
supervised approach which is image classification:

Example of face detection and recognition:

(Murphy family, photo
taken 5 August 2010).

2. Unsupervised Machine Learning

Unsupervised learning is arguably more typical of human and animal
learning. It is also more widely applicable than supervised learning, uses
machine learning algorithms to analyze and cluster unlabeled datasets.
unsupervised learning, where we are just given output data, without any
inputs. The goal is to discover “interesting structure” in the data; this is
sometimes called knowledge discovery. This method’s ability to discover
similarities and differences in information make it ideal for exploratory
data analysis, cross-selling strategies, customer segmentation, and image
and pattern recognition. It’s also used to reduce the number of features in a
model through the process of dimensionality reduction. Principal
component analysis (PCA) and singular value decomposition (SVD) are
two common approaches for this. Other algorithms used in unsupervised
learning include neural networks, k-means clustering, and probabilistic
clustering methods.
Geoff Hinton, who is a famous professor of ML at the University of Toronto,
has said: When we’re learning to see, nobody’s telling us what the right
answers are—we just look. Every so often, your mother says “that’s a dog”,
but that’s very little information. You’d be lucky if you got a few bits of
information—even one bit per second—that way. The brain’s visual
systemhas1014neural connections. And you only live for109 seconds. So
it’s no use learning one bit per second. You need more like105 bits per
second. And there’s only one place you can get that much information: from
the input itself.—Geoffrey Hinton, 1996(quotedin(Gorder2006)).
Below we lists some canonical examples of unsupervised learning.
I. Discoveringclusters
II. Discovering latent factors
III. Discovering graph structure
IV. Matrix completion
V. Image inpainting
VI. Collaborative filtering
VII. Market basket analysis

(Machine Learning A Probabilistic Perspective Kevin P. Murphy & IBM)

Major Applications of ML

Machine learning is used in a wide variety of industries — not just in the

tech-heavy companies you might imagine when you think about someone
manipulating an enormous data set. Here, we will discuss machine learning
use cases by industry, providing some of the different ways that this tool is
being used today by a variety of people and companies.
 ▪ Machine Learning Applications in Finance
In the financial services sector, analysts use machine learning to automate
trading activities, detect fraud, and provide financial advising services
to their clients. Algorithmic trading requires traders to build mathematical
models that can monitor news feeds and trading trends to predict a rise or
fall in security prices. Finance companies also use machine learning to
detect fraudulent activity by comparing transactions against other existing
data points.

▪ Machine Learning Applications in Business

Machine learning offers businesses an extensive number of ways to boost
their effectiveness, efficiency, and offerings. Chatbots, for example, allow
businesses to provide faster, more flexible customer service without
employing a call center or making customers wait on hold for the next
available representative.

▪ Machine Learning Applications in Genetics and Genomics

The human genome is one of the largest data sets ever studied. Humans
contain over 20,000 different genes, each of which has potential for
variation. Machine learning allows researchers to better understand
different genetic traits and abnormalities as they analyze and understand
vast data sets.

▪ Machine Learning Applications in Healthcare

When applied to healthcare, machine learning can help hospitals and their
staff make administrative processes more efficient and streamlined,
personalize medical treatments, and better understand and track infectious
diseases.

▪ Machine Learning Applications in Retail

As we’ve already mentioned, machine learning can be incredibly helpful in
understanding and decreasing customer churn (i.e., the rate at which a
business loses customers each year), which is a large point of focus for many
retail companies.

▪ Machine Learning Applications in Education

Machine learning can help educational institutions on both process-based
tasks and more student-focused initiatives. Statistical models can help
understand student progress and needs, while scheduling algorithms can
help create more efficient and streamlined schedules for institutions of all
sizes.
 ▪ Farming
Tractor manufacturers are working on developing deep learning algorithms
to classify and assess the quality of harvested grains and make automatic
adjustment systems for the harvester. (Perry, 2019)

▪ Cyberthreat detection
Machine learning's capacity to analyze complex patterns within high
volumes of activities to both determine normal behaviors and identify
anomalies also makes it a powerful tool for detecting cyberthreats.
(The Fu Foundation School of Engineering and Applied Science, internet resources)
 The benefits of ML to our new AI based world

(Photo downloaded from Columbia Engineering Boot Camps)

Conclusion

Considering all aspects, it is difficult to say how much benefit it will bring
to humans, but it will play a very effective role in the business sector.
Another point is that customers should have at least some knowledge
about its functionality; otherwise, its harmful aspects may reach their
peak.