Chapter -1 Overview of ML
What is Machine Learning?
Machine learning is a growing technology which enables computers to learn automatically
from past data
we have computers or machines which work on our instructions. But can a machine also
learn from experiences or past data like a human does?
2
What is Machine Learning?
The term machine learning was first introduced by Arthur Samuel in 1959
He defined machine learning as “the field of study that gives computers the ability to
learn without being explicitly programmed.”
However, there is no universally accepted definition. Different authors define the term
differently.
Tom Mitchell (1998):
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.
3
What is Machine Learning?
Breaking Down
Experience (E): This refers to the data or examples the system is exposed to
Tasks (T): These are the specific activities or problems that the machine is learning to perform or
solve.
Performance Measure (P): This is the metric used to evaluate how well the machine is doing in
performing tasks from T. accuracy, error rate, or another relevant metrics.
Example
Handwriting recognition learning problem,
Task T: Recognizing and classifying handwritten words within images,
Performance P: Percent of words correctly classified,
Experience E: A dataset of handwritten words
4
What is Machine Learning?
Email Spam Detection Problem:
Task T: classify email as spam or not-spam
Performance P: number of correctly predicted spams
Experience E: A dataset of labelled emails
Other Def:
Machine learning is programming computers to optimize a performance criterion using
example data or past experience.
It is a branch of Artificial Intelligence and computer science that helps build a model
based on training data and make predictions and decisions without being constantly
programmed.
5
How ML Help Us?
It is very hard to write programs that solve problems like recognizing a face.
✔
We don’t know what program to write because we don’t know how our brain does it.
✔
Even if we had a good idea about how to do it, the program might be horrendously
complicated.
A. Instead of writing a program by hand, we collect lots of examples that specify the
correct output for a given input.
B. A machine learning algorithm then takes these examples and produces a program
that does the job.
6
How ML works?
●
A Machine Learning system learns from historical data, builds the prediction
models(Using Algorithms), and whenever it receives new data, predicts the
output for it.
●
The accuracy of predicted output depends upon the amount of data, as the huge
amount of data helps to build a better model
7
Features of ML
uses data to detect various patterns in a given dataset.
It can learn from past data and improve automatically.
It is a data-driven technology
is much similar to data mining as it also deals with the huge amount of the data.
8
Need of ML
The need for machine learning is increasing day by day. The reason behind
the need for machine learning is that it is capable of doing tasks that are too
complex for a person to implement directly.
Rapid increment in the production of data
Solving complex problems, which are difficult for a human
Decision making needs in various sector including finance
Finding hidden patterns and extracting useful information from data.
9
History of ML
Before some years , machine learning was science fiction, but today it is the
part of our daily life.
Has old history
10
History of ML
The early history of Machine Learning (Pre-1940):
✔ 1834: Charles Babbage, the father of the computer, conceived a device that could be
programmed with punch cards.
✔ 1936: Alan Turing gave a theory that how a machine can determine and execute a set of
instructions.
The era of stored program computers:
1940: the first manually operated computer, "ENIAC" was invented, which was the first
electronic general-purpose computer.
1950, the scientists started applying their idea to work and analyzed how human neurons
might work.
Computer Machinery and Intelligence:
1950: Alan Turing published a seminal paper, "Computer Machinery and Intelligence," on the
topic of artificial intelligence. In his paper, he asked, "Can machines think?"
Machine intelligence in Games:
1952: Arthur Samuel,created a program that helped an IBM computer to play a checkers game
11
History of ML...
The first "AI" winter:
The duration of 1974 to 1980 was the tough time for AI and ML researchers,
and this duration was called as AI winter.
Machine Learning from theory to reality
✔
1959: the first neural network was applied to a real-world problem to
remove echoes over phone lines
✔
1985: Terry Sejnowski and Charles Rosenberg invented a neural
network NETtalk, which was able to teach itself how to correctly
pronounce 20,000 words in one week.
✔
1997: The IBM's Deep blue intelligent computer won the chess game
against the chess expert Garry Kasparov, and it became the first computer
which had beaten a human chess expert.
12
History of ML ...
Machine Learning at 21st century
Deep learning
Deep Face
Chat boat technologies and others invented
Machine Learning at Present
Now machine learning has got a great advancement in its research, and it is
present everywhere around us, such as self-driving
cars, Catboats, recommender system,Various AI technologies and many more
13
Classification of ML
In general, machine learning algorithms can be classified into three
types.
Supervised learning
Unsupervised learning
Reinforcement learning
14
Supervised Learning
Supervised learning involves training an algorithm on a labeled
dataset, where input data is paired with corresponding output labels.
The goal is to learn a mapping from input to output based on
provided labelled examples.
●
A “supervised learning” is so called because the process of an
algorithm learning from the training dataset can be thought of as a
teacher supervising the learning process
15
Supervised Learning
16
Supervised Learning
Types of Problem To Solve
Classification:
✔
Predicting whether an email is spam or not based on features like subject,
sender, and content.
Regression:
✔
Predicting the price of a house based on features like square footage, location,
and the number of bedrooms.
Health care, financial, sentiment....
17
Unsupervised Learning
Unsupervised learning involves training a model on not labeled, classified, or
categorized, data and the algorithm needs to act on that data without any supervision,
The algorithm aims to explore the inherent structure in the data.
models itself find the hidden patterns and insights from the given data.
The goal is to restructure the input data into new features or a group of objects with
similar patterns.
It can be compared to learning which takes place in the human brain while learning new
things
18
Unsupervised Learning
19
Unsupervised ML Problem Types
A. Clustering:
Grouping similar customers based on purchasing behaviour without predefined
categories.
•Market Segmentation: Identifying distinct customer segments for targeted
marketing campaigns.
•Anomaly Detection: Detecting unusual patterns in network traffic for
cybersecurity.
B. Association :
find associations and patterns in the data ex : dimensionality reduction
More examples
20
Re-Inforcement Learning
Reinforcement learning involves an agent learning to make decisions by
interacting with an environment.
The agent receives feedback in the form of rewards or penalties based on the
actions it takes.
The agent learns automatically with these feedbacks and improves its
performance
Reinforcement learning is applicable when an agent needs to learn a sequence
of actions to achieve a goal in a dynamic environment, receiving feedback to
guide its learning process.
21
Re-Inforcement Learning
22
Re-Inforcement Learning ex:
Controlling A Walking Robot
Agent:
The program controlling a walking robot.
Environment:
The real world.
Action:
One out of four moves (1) forward; (2) backward; (3) left; and (4) right.
Reward:
Positive when it approaches the target destination; negative when it wastes
time, goes in the wrong direction or falls down.
Example 2: The robotic dog, which automatically learns the movement of his
arms,
23
Challenges In Machine Learning
●
Machine learning projects encounter challenges such as
●
data quality issues,
●
insufficient data
●
the need for robust model evaluation.
Limitations of current machine learning approaches:
While powerful, current ML approaches have limitations, including
interpretability issues, susceptibility to bias, and challenges in handling
complex, unstructured data.
Understanding these limitations is crucial for responsible and effective deployment.
24
Application Areas of ML
25
Application areas of ML
A. Image Recognition
● It is used to identify objects, persons, places, digital images
●
Facebook Deepface - Automatic friend tagging suggestion
B. Speech Recognition
● process of converting voice instructions into text, and it is also known as "Speech to text", or
"Computer speech recognition."
● Google Search by voice,
Example: Google Assistant, Siri, Cortana(Microsoft), and Alexa(amazon)
C. Traffic prediction : google map
D. Product recommendations:
●
e-commerce and entertainment companies such as Amazon, Netflix
26
Application areas of ML
E. Self-driving cars
Tesla, the most popular car manufacturing company is working on self-driving car
F. Email Spam and Malware Filtering
● When we receive new email, it is filtered automatically as important, normal, and spam
G. Virtual Personal Assistant
● Help us using voice instructions such as Play music, call someone, Open an email, Scheduling
an appointment,
● various virtual personal assistants such as Google assistant, Alexa, Cortana, Siri.
27
Machine learning Life cycle
Machine learning life cycle is a cyclic process to build an efficient machine learning project
Gathering Data
Data preparation
Data Wrangling
Analyse Data
Train the model
Test the model
Deployment
28
Steps of ML
1. Gathering Data
➔ Identify various data sources ,Collect data and Integrate the data obtained from different
sources-> datasets
➔
Open data sources: UCI ML repository,kaggle,OpenML,google and Microsoft Research Open Data
➔ [Link] Preparation: a step where we put our data into a suitable place and prepare it to use in
our machine learning training ex: Data exploration,EDA
➔
[Link] Wrangling : is the process of cleaning and converting raw data into a useable format.
Missing Values ,Duplicate data ,Invalid data , Noise removal
29
Steps of ML
4. Data Analysis : Now the cleaned and prepared data is passed on to the analysis step. This
step involves: Selection of analytical techniques ,Building models ,Review the result
5. Train Model : -in this step we train our model to improve its performance for better
outcome of the problem.
6. Test Model : once trained we check for the accuracy of our model by providing a test
dataset to it.
7. Deployment
we deploy the model in the real-world system.
30
Data Science
What is Data science
"A field of deep study of data that includes extracting useful insights from the data, and
processing that information using different tools, statistical models, and Machine learning
algorithms.
●
Concept that is used to handle big data that includes data cleaning, data preparation, data
analysis, and data visualization.
31
Skills needed to be Data Scientist
An excellent programming knowledge of Python, R, SAS, or Scala.
Experience in SQL database Coding.
Knowledge of Machine Learning Algorithms.
Deep Knowledge of Statistics concepts.
Data Mining, cleaning, and visualizing skills.
Skills to use Big data tools such as Hadoop.
32
Data Science VS ML
33
AI vs ML vs DL
34
AI,ML and DL Summary
Artificial intelligence is the overarching field focused on creating intelligent
systems.
●
Machine learning is a subset of artificial intelligence that emphasizes learning
from data to make predictions or decisions.
●
Deep learning is a subfield of machine learning that uses neural networks with
multiple layers to learn hierarchical representations of data.
35
End of Chapter-one
36
