Chapter One

Introduction to Machine Learning

By: Petros Abebe Mizan Tepi University Jun, 2023 1/21

What is learning?

Learning is the relatively permanent

change in a person’s knowledge or
behavior due to experience.
It is the acquisition of information,
knowledge, and skills.
To gain knowledge or understanding
of, or skill by study, instruction or
experience
Learning a set of new facts.
Learning HOW to do something .
Improving ability of something
already learned.

By: Petros Abebe Mizan Tepi University Jun, 2023 2/21

What is Machine Learning?

Machine Learning is said as a subset of artificial intelligence

that is mainly concerned with the development of algorithms
which allow a computer to learn from the data and past
experiences on their own.
In traditional programming, humans write explicit instructions
for a computer to perform specific tasks. However, in machine
learning, the computer learns from data and experiences to
improve its performance on a particular task.
Machine learning algorithms are trained over examples through
which they learn from past experiences and also analyze the
historical data.

By: Petros Abebe Mizan Tepi University Jun, 2023 3/21

Cont . . .

The term machine learning was first introduced by Arthur

Samuel in 1950s.
”Machine learning enables a machine to automatically learn
from data, improve performance from experiences, and
predict things without being explicitly programmed.”
Tom M. Mitchell 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 the experience E.

By: Petros Abebe Mizan Tepi University Jun, 2023 4/21

Why machine learning?

The goal of machine learning is to develop methods that can

automatically detect patterns in data, and then to use the
uncovered patterns to predict future data or other outcomes of
interest
Machine learning (ML) has become a central focus in
technology and science due to its transformative potential across
various industries and applications.
Automation of Tasks
Data Analysis and Insights
Enhanced Decision-Making
Innovation in Various Fields: (Healthcare, Autonomous Systems,
Natural Language Processing)
Addressing Complex Problems

By: Petros Abebe Mizan Tepi University Jun, 2023 5/21

Components of Machine Learning

Machine learning derives insightful information from large

volumes of data by leveraging algorithms to identify patterns and
learn in an iterative process.

By: Petros Abebe Mizan Tepi University Jun, 2023 6/21

Components of Machine Learning

Data
Crucial to machine learning, data is the raw information that is
processed and analyzed by algorithms.
It can be structured or unstructured and comes in various forms
such as text, images, numbers, and more.
Quality and quantity of data significantly influence the
performance of a machine learning model.
Data can come in various forms such as structured data (e.g.
databases), unstructured data (e.g. text, images), and semi
structured data (e.g.JSON files).

By: Petros Abebe Mizan Tepi University Jun, 2023 7/21

Components of Machine Learning
Algorithm
An algorithm is a set of mathematical instructions that a machine
learning model uses to learn patterns from data.
Algorithms are the heart of machine learning and dictate how
data is transformed into actionable knowledge.
Choosing the right algorithm depends on the type of problem
being solved and the nature of the data.

Model
In machine learning, a model represents what an algorithm has
learned from the training data.
It is the output that you get after training an algorithm with data,
and it is this model that is used to make predictions or decisions.
Models need to be validated and tested to ensure they generalize
well to unseen data.
By: Petros Abebe Mizan Tepi University Jun, 2023 8/21
Types of Machine Learning

By: Petros Abebe Mizan Tepi University Jun, 2023 9/21

Supervised Learning Algorithms

Supervised Learning is a category of

machine learning that uses labeled datasets
to train algorithms to predict outcomes and
recognize patterns.
The algorithm learns on a labeled datasets,
to generate expected predictions for the
response to new data.
Labeled data refers to a dataset that
includes both the input data and the
corresponding correct output.
The goal is for the model to learn the
mapping from inputs to outputs so that it
can predict the output for new, unseen data.

By: Petros Abebe Mizan Tepi University Jun, 2023 10/21

By: Petros Abebe Mizan Tepi University Jun, 2023 11/21
Supervised learning Tasks

Classification
Classification is the process of categorizing data or objects into
predefined classes or categories based on their features or
attributes.
Classification is the process of categorizing data or objects into
predefined classes or categories based on their features or
attributes.
Classification involves predicting a categorical or discrete
output variable based on a set of input features.
The goal of classification is to learn a model that can accurately
assign new input data to one of several predefined categories or
classes.

By: Petros Abebe Mizan Tepi University Jun, 2023 12/21

Supervised Learning Tasks

Regression
Regression involves predicting a continuous or numerical output
variable based on a set of input features.
The goal of regression is to learn a model that can accurately
predict a numerical value for new input data.
Regression is used to predict a continuous output based on
historical input data only.
Some common applications of regression include predicting
stock prices, estimating housing prices, and forecasting weather.

By: Petros Abebe Mizan Tepi University Jun, 2023 13/21

Unsupervised Learning

A type of machine learning where the algorithm is trained on

unlabeled data, meaning that the data does not have pre-existing
labels or outcomes.
The goal of unsupervised learning is to uncover hidden patterns,
structures, or relationships within the data.
It is enabled to predict the output without any supervision.
The algorithm learns on an unlabeled datasets and tries to make
sense by extracting features, co-occurrence, and underlying
patterns on its own.

By: Petros Abebe Mizan Tepi University Jun, 2023 14/21

By: Petros Abebe Mizan Tepi University Jun, 2023 15/21
Unsupervised Learning Tasks
Clustering
Clustering: is a technique used to group similar data points
together based on their features or characteristics.
The goal is to partition the data into subsets, or clusters, such that
points within a cluster are more similar to each other than to
points in other clusters.
Some clustering algorithms are K-Means clustering, Hierarchical
Clustering, DBSCAN Clustering

Dimensionality reduction
It is a technique used to reduce the number of features in a
dataset while retaining as much of the original information as
possible.
This is useful for visualizing high-dimensional data or for
reducing the computational complexity of a machine learning
model.
By: Petros Abebe Mizan Tepi University Jun, 2023 16/21
Reinforcement Learning

Reinforcement Learning is a subset of machine learning that

enables a component of the system (called agent):
to learn in a simulated, interactive virtual environment by trial
and error
reinforcing the results using a reward-punishment system that is
generated using feedback from its own actions and experiences.
The agent receives feedback in the form of rewards or
punishments based on its actions, and it learns to maximize its
rewards over time.
Reinforcement learning is a feedback-based process.
Reinforcement learning is commonly used in applications such
as game playing, robotics, and autonomous driving.

By: Petros Abebe Mizan Tepi University Jun, 2023 17/21

Application of Machine Learning

By: Petros Abebe Mizan Tepi University Jun, 2023 18/21

Cont . . .

Image recognition: is one of the most common applications of

machine learning. It is used to identify objects, persons, places,
digital images, etc.
Product recommendation: Machine learning is widely used by
various e-commerce and entertainment companies such as
Amazon, Netflix, etc., for product recommendation to the user.
Medical Diagnosis: Machine learning is used for diseases
diagnoses. With this, medical technology is growing very fast
and able to build 3D models that can predict the exact position of
lesions in the brain.
Language Translation: machine learning helps us by
converting the text into our known languages. Google’s GNMT
(Google Neural Machine Translation) provide this feature, which
is a Neural Machine Learning that translates the text into our
familiar language, and it called as automatic translation.

By: Petros Abebe Mizan Tepi University Jun, 2023 19/21

Machine Learning Libraries

NumPy
NumPy is the fundamental package for scientific computing in
Python.
The name is an acronym for ”Numeric Python” or ”Numerical
Python”.
Python with powerful data structures, implementing
multi-dimensional arrays and matrices.

matplotlib
These libraries allow users to plot much cleaner and
sophisticated graphs.
Matplotlib is an visualization library in Python for 2D plots of
arrays.

By: Petros Abebe Mizan Tepi University Jun, 2023 20/21

Machine Learning Libraries

pandas
pandas is a Python package that provides fast, flexible, and
expressive data structures designed to make working with
”relational” or ”labeled” data both easy and intuitive.
Pandas is a Python library that is mainly used for data analysis.

Scikit-learn
Scikit-learn is a Python library which is used for classical
machine learning algorithms.
Scikit-learn is popular in Machine learning developers as it
supports supervised and unsupervised learning algorithms.
This library can also be used for data-analysis, and data-mining
process.

By: Petros Abebe Mizan Tepi University Jun, 2023 21/21

By: Petros Abebe Mizan Tepi University Jun, 2023 22/21