This document provides an overview of machine learning concepts including supervised and unsupervised learning algorithms. It discusses regression, classification, and clustering techniques. Specifically, it covers: - Supervised learning algorithms like linear regression, logistic regression, neural networks, and support vector machines for both regression and classification problems. - Unsupervised learning algorithms like clustering and principal component analysis (PCA) to derive structure from unlabeled data. - Examples of applying machine learning include predicting housing prices with regression, classifying medical images, and grouping similar documents or genes with clustering.

1. CCS 3081
Instructor: Dr. Emad Nabil
Machine Learning

2. Some of the
Machine learning
algorithms
Supervised
Learning
Regression
Linear regression with one variable
linear regression with multiple vars
Regularization and Overfitting
Classification
logistic regression
Regularization and Overfitting
Neural networks
Support vector machine
Naive Bayes
Decision Tree
Unsupervised
Learning
clustering
PCA
K-Means

3. Agenda
 What is machine learning?
 When Do We Use Machine Learning?
 Supervised Learning
 Regression
 Classification
 Unsupervised Learning
 clustering

4. Tools
 Python interpreter : https://www.anaconda.com/products/individual
 Python IDE: https://www.jetbrains.com/pycharm/download/#section=windows

5. What is machine learning?
Machine learning: "Field of study that gives computers the ability to
learn without being explicitly programmed“
- Arthur Samuel (1959)

6. What is machine learning?

7. When Do We Use Machine Learning?
ML is used when:
Human expertise does not exist (navigating on Mars)

8. When Do We Use Machine Learning?

9. Examples of tasks can be solved by using ML
• Recognizing patterns:
• Facial identities
• Facial expressions
• Handwritten recognition
Ahmed
Mohamed
Sad
Happy

10. Examples of tasks can be solved by using ML
• Recognizing patterns:
• Medical images
• Recognizing anomalies:
– Unusual credit card transactions
• Prediction:
– Future stock prices
COVID-19
Free
Noraml withdraw
fraudulent
withdrawal

11. Main Machine learning algorithms
Supervised
Learning
Regression Classification
Unsupervised
Learning
clustering

12. Supervised
Learning
Regression Classification
Supervised Learning

13.  In supervised learning, we are given a data set and already know what our correct output should look like.
 In a regression problem, we are trying to predict results within a continuous output.
 In a classification problem, we are instead trying to predict results in a discrete output or discrete categories.
Supervised Learning

14. x1 x2 x3 Y (has disease or not)
83 150 Tech. 1

16. 0
100
200
300
400
0 500 1000 1500 2000 2500
Supervised Learning -Regression problem
Housing price prediction.
Price ($) in 1000’s
Size in feet2
Regression: Predict continuous valued output (price)
 What approaches can we use to solve this?
 Straight line through data Maybe $150 000
 Second order polynomial Maybe $200 000
 One thing we discuss later how to chose straight or curved line?

17. Supervised Learning -classification problem
• Case with cancer  case with no caser 
• Given the data below how to determine a new input case
is malignant or benign case
 age and tumor size are called attributes or features, You might have many features to consider like:
• Clump thickness
• Uniformity of cell size
• Uniformity of cell shape
 The most exciting algorithms can deal with an infinite number of features, like SVM, will see later.
x
x

18. Supervised Learning
Example 1:
Example 2 :
Example 3:
Supervised
Learning
Regression Classification
Age ?
Cancer ?
Price ?

19. Main Machine learning algorithms
Supervised
Learning
Regression Classification
Unsupervised
Learning
clustering

20. Unsupervised Learning
Unsupervised
Learning
clustering

21. Unsupervised Learning
 Unsupervised learning allows us to approach problems with no idea what our results should look like.
 We can derive structure from data, by clustering the data based on relationships among the variables in the data.
 With unsupervised learning there is no feedback based on the prediction results, i.e., there is no teacher to correct you.

22. Unsupervised Learning applications
Example: 1
Clustering: Take a collection of 1000 essays written documents , and find a way to automatically group these essays into a
small number that are somehow similar or related by different variables, such as: sentence length, or page count.

23. Unsupervised Learning applications
Example:2
Clustering: 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, like :
• lifespan,
• location,
• roles

24. Social network analysis
Unsupervised Learning applications

25. Market segmentation
Unsupervised Learning applications