Scribd
Upload
50 views47 pages

ML Lec 1

The examples given are: 1. Given data about the size of houses on the real estate market, try to predict their price. - This is regression as it tries to predict a continuous numerical value (price of houses) based on input data. 2. Given a patient with a tumor, we have to predict whether the tumor is malignant or benign. - This is classification as it tries to predict a discrete class/category (malignant or benign) based on input data. 3. Building a system that predicts whether the house "sells for more or less than the asking price." - This is classification as it tries to predict a discrete class (sells for more or less) based on input data.

Uploaded by

Areej Ehsan
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
50 views47 pages

ML Lec 1

The examples given are: 1. Given data about the size of houses on the real estate market, try to predict their price. - This is regression as it tries to predict a continuous numerical value (price of houses) based on input data. 2. Given a patient with a tumor, we have to predict whether the tumor is malignant or benign. - This is classification as it tries to predict a discrete class/category (malignant or benign) based on input data. 3. Building a system that predicts whether the house "sells for more or less than the asking price." - This is classification as it tries to predict a discrete class (sells for more or less) based on input data.

Uploaded by

Areej Ehsan
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 47

Machine Learning

Lecture 1
Dr. Muhammad Kamran Malik
What is Machine Learning
What is Machine Learning
• A subfield of computer science and AI
• Learn, make decisions and prediction from data.
• Enables computers to act and make data–driven decision rather than
being explicitly programmed to carry out a certain task.
What is Machine Learning

• The field of study that gives computers the ability to


learn without being explicitly programmed. (Arthur
Samuel-1959)
What is Machine Learning

• 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.
(Tom Mitchell-1998)
• Playing checkers
• E = the experience of playing many games of checkers
• T = the task of playing checkers.
• P = the probability that the program will win the next
game.
What is Machine Learning
Example: playing checkers

• "A computer program is said to learn from experience


E with respect to some task T and some performance
measure P, if its performance on T, as measured by P,
improves with experience E."
• Suppose your email program watches which emails
you do or do not mark as spam, and based on that
learns how to better filter spam. What is the task T in
this setting?
Example: playing checkers

• "A computer program is said to learn from experience E with


respect to some task T and some performance measure P, if its
performance on T, as measured by P, improves with experience
E."
• Suppose your email program watches which emails you do or
do not mark as spam, and based on that learns how to better
filter spam. What is the task T in this setting?
• Classify emails as spam or not spam.
• Watching you label emails as spam or not spam.
• The number (or fraction) of emails correctly classified as spam/not
spam.
• None of the above, this is not a machine learning algorithm.
When Machine Learning?

• Learning is used when:


• Human expertise does not exist (navigating on Mars),

• Humans are unable to explain their expertise (speech


recognition, time series analysis)

• Solution changes in time (routing on a computer network)

• Solution needs to be adapted to particular cases (user


biometrics)

• Data is Big Data (Volume, Velocity, Variety)


Q&A
Machine Learning Applications
Machine Learning Applications (Virtual Personal Assistant)
Machine Learning Applications (Virtual Personal Assistant)
Machine Learning Applications (Traffic Prediction)
Machine Learning Applications (Email Spam Filtering)
Machine Learning Applications (chatbots)
Machine Learning Applications (Search Engine Result Refining)
Machine Learning Applications (Online Fraud Detection)
Machine Learning Applications (Stock Market Trading)
Machine Learning Applications (Assistive Medical Technology)
Machine Learning Applications (Automatic Translation)
Machine Learning Applications (Automatic Friend Tagging suggestions)
Machine Learning Applications (Alt Tags)
Machine Learning Applications (Recommendation Systems)
Machine Learning Applications

• Autonomous Cars (TESLA)


• Face Recognition app (Iphone)
• Moley (The Robotic Chef)
• KUKA (Robotic Automation)
Q&A
Machine Learning Process
Lifecycle
Machine Learning Process Lifecycle
Get Data
Cleaning and preparing (Features)
Train Model (Model Building – a search problem)

• Which type of model is most suitable for this data?

(a) (b)

Dataset - 1 Dataset - 2 (c) (d)


Test Model (Evaluation)
• Given prediction results, which model is better?
Model-1 Predicted

100 50 50

Actual 1000 100 900

Accuracy = 950 / 1100 = 86.4% Model-2 Predicted

100 0 100

Actual
1000 0 1000

Accuracy = 1000 / 1100 = 91% 33


Q&A
Types of Machine Learning Algo
Types of Machine Learning Algorithms
Regression
Supervised Learning

Machine Learning
[Task Driven]
Classification

Clustering
Unsupervised Learning
[Data Driven]
Dimensionality
Reduction

Fully observable
environment
Reinforcement Learning
[Environment Driven]
Partially observable
environment
Supervised Learning
Supervised Learning
Unsupervised Learning
Reinforcement Learning
Machine Learning Extended
Q&A
Clustering / Classification /
Regression
Clustering / Classification / Regression ??

• Given data about the size of houses on the real estate market, try to
predict their price.
• Given a picture of a person, we have to predict their age on the basis of
the given picture
• Given a patient with a tumor, we have to predict whether the tumor is
malignant or benign.
• Building a system that predicts whether the house "sells for more or less
than the asking price."
• Take a collection of 1,000,000 different genes, and find a way to
automatically group these genes into groups that are somehow similar
or related by different variables, such as lifespan, location, roles, and so
on.
• The "Cocktail Party Algorithm",  This is unsupervised
allows you to find structure in a chaotic (clustering)
environment. (i.e. identifying individual
voices and music from a mesh of
sounds at a cocktail party).
 This is unsupervised
(clustering)
Eg: recommended systems , filtering news
feed

You might also like