2. Friendly Introduction to Machine Learning
Learn from experience Learn from data Follow Instructions
3. What is Machine Learning?
Machine Learning is the science of getting computers to act without being
explicitly programmed.
Computer
Data
Program
Output
Computer
Data
Output
Program
Traditional Programming:
Machine Learning:
4. Why is it trending now?
Availability of many different kinds of data.
Greater computational power.
Exponential decrease in the price of powerful computing
resources.
Train computers to do things that are very difficult to
program.
5. Machine Learning Methods
ML Methods
Supervised
Learning
Unsupervised
Learning
ClassificationRegression Clustering Association
11. Support Vector Machine
Vladimir Vapnik laid most of the groundwork for SVM
while working on his PhD thesis in the Soviet Union in
1960s.
A supervised Machine Learning algorithm
Used for both classification and regression
It’s a binary classifier
12. Support Vector Machine
.
SVM is a classifier method that performs classification tasks by constructing
hyperplanes.
14. Identifying the right Hyperplane?
Margin
Maximum perpendicular distance between the nearest data point and hyperplane -
Margin
15. How to compute MARGIN?
Consider w perpendicular to median.
Consider u, which we would like to classify.
Decision rule:
w • u ≥ c ⇒ u is +, or
w • u + b ≥ 0 ⇒ u is +,
where c= -b.
w • x++ b ≥ 1 (+ve sample)
w • x-+ b ≤ −1 (-ve sample)
16. Introduce yi = 1 for + data and yi = -1 for -
data.
yi(x • w + b) ≥ 1, or
yi(x • w + b) − 1 ≥ 0,
Margin =
How to compute MARGIN?
24. What if the data is not linearly separable?
Idea : Separable in higher dimension
X
Y
Z=X2 +Y2
X
Z
Y
25. Where does SVM get its name?
Separating plane is usually determined by only a handful
of data points.
The points that help determine the hyperplane are called
Support Vectors.
The hyperplane itself is a classifying machine.
26. WEKA: A Machine
Learning tool
Waikato Environment for Knowledge
Analysis
Open source software tool
Developed at The University of
Waikato
Collection of visualization tools and
algorithms