1
00:00:04,410 --> 00:00:06,701
So in this module we're
gonna talk about regression.

2
00:00:06,701 --> 00:00:09,486
Which is one of the most widely
used statistical tools outs there.

3
00:00:09,486 --> 00:00:11,650
And the idea is really simple.

4
00:00:11,650 --> 00:00:16,070
So we have some set of features and
we wanna model how our observations that

5
00:00:16,070 --> 00:00:22,200
are associated with these features change
as we change the values of the features.

6
00:00:22,200 --> 00:00:25,670
And we're gonna ground our
conversation in a case study

7
00:00:25,670 --> 00:00:28,640
of trying to predict house values.

8
00:00:28,640 --> 00:00:32,748
So here we can imagine that a house has
some set of features like what's the size

9
00:00:32,748 --> 00:00:36,420
of the house, how many bedrooms does
it have, number of bathrooms and

10
00:00:36,420 --> 00:00:37,840
the list goes on and on.

11
00:00:37,840 --> 00:00:42,110
And the observation that we have is
what's the value of the house or

12
00:00:42,110 --> 00:00:43,770
the house sales price.

13
00:00:43,770 --> 00:00:47,120
But the tools of regression go

14
00:00:47,120 --> 00:00:50,120
much beyond just thinking
about doing prediction tasks.

15
00:00:50,120 --> 00:00:52,950
So as we're gonna see in
the classification course,

16
00:00:52,950 --> 00:00:56,230
we can use regression tools for
classification.

17
00:00:56,230 --> 00:00:58,830
So for example,
let's say we have an email and

18
00:00:58,830 --> 00:01:01,450
we wanna classify whether it's spam or
not spam.

19
00:01:01,450 --> 00:01:06,120
Well, that email we can think of having
features about the text of the e-mail,

20
00:01:06,120 --> 00:01:10,190
that's indicative of whether
that e-mail is spam or not spam.

21
00:01:10,190 --> 00:01:13,225
And in addition we can think
about using regression for

22
00:01:13,225 --> 00:01:15,580
analyze the importance of
the features themselves.

23
00:01:15,580 --> 00:01:18,884
And we're gonna talk more about
that in the regression course.

24
00:01:18,884 --> 00:01:22,289
[MUSIC]