1
00:00:01,040 --> 00:00:07,760
So in this course, we are going to discuss the out of box machine learning technique called support

2
00:00:07,760 --> 00:00:08,660
vector machines.

3
00:00:10,930 --> 00:00:18,460
And the beginning of the course, we will discuss maximal marginal, classify this maximal margin,

4
00:00:18,460 --> 00:00:26,710
classify it is a very simple and intuitive classifier and support vector machine is a generalization

5
00:00:26,830 --> 00:00:28,300
of this classifier only.

6
00:00:30,180 --> 00:00:36,750
However, this classifier can only be applied when classes are linearly separable.

7
00:00:38,190 --> 00:00:42,690
Therefore, the application of this classifier are severely limited.

8
00:00:43,890 --> 00:00:50,610
Next, we will discuss support recook classifier, which is an extension of the maximal marginal classifier

9
00:00:51,600 --> 00:00:54,630
and can be applied in a broader range of cases.

10
00:00:56,160 --> 00:01:02,160
Lastly, we will cover support vector machines, which is a further extension of the support vector

11
00:01:02,320 --> 00:01:02,790
classifier.

12
00:01:04,710 --> 00:01:09,240
We will see that how SBM accommodate non-linear class boundaries.

13
00:01:09,630 --> 00:01:17,550
Usually Astrium is meant to handle binary classification, but later on we will also see how to extend

14
00:01:17,850 --> 00:01:21,330
support vector machines to the case of more than two classes.

15
00:01:22,770 --> 00:01:28,830
Oftentimes these three terms are used interchangeably, but there is a difference between these three

16
00:01:29,190 --> 00:01:33,960
types of classifiers and we will note what is the difference and how these three are different.