﻿1
00:00:00,266 --> 00:00:02,600
Hello and welcome to this art tutorial.

2
00:00:02,600 --> 00:00:04,766
So now, in this new
step of the cleaning process,

3
00:00:04,766 --> 00:00:07,833
we will remove all the punctuations
in the reviews of the corpus.

4
00:00:08,100 --> 00:00:10,600
And it's going to be as simple as before.

5
00:00:10,600 --> 00:00:13,633
We will copy this line, paste

6
00:00:13,666 --> 00:00:17,100
it below
and instead of remove numbers here

7
00:00:17,133 --> 00:00:20,133
we will input remove punctuation.

8
00:00:20,966 --> 00:00:22,266
As simple as that.

9
00:00:22,266 --> 00:00:23,600
We don't need anything else.

10
00:00:23,600 --> 00:00:28,766
And that's ready to remove any punctuation
and all the reviews of the corpus.

11
00:00:29,200 --> 00:00:30,433
All right so let's check it out.

12
00:00:30,433 --> 00:00:32,966
We can actually check it out
with the first review.

13
00:00:32,966 --> 00:00:35,966
Well it's really all that love this place.

14
00:00:36,033 --> 00:00:40,000
And so what we're supposed to obtain
after applying this remove

15
00:00:40,000 --> 00:00:43,866
punctuation function and all the reviews
of the corporate through the tmp function.

16
00:00:44,300 --> 00:00:47,300
Well, the three little dots
are supposed to disappear.

17
00:00:47,566 --> 00:00:49,333
All right. So let's check it out.

18
00:00:49,333 --> 00:00:52,833
So remember in the actual version
of the corpus right now,

19
00:00:52,833 --> 00:00:56,800
the first review is wow, thrilled
that I love this place.

20
00:00:56,966 --> 00:01:00,066
So now let's select this new line of code,

21
00:01:00,633 --> 00:01:05,000
execute new corpus of reviews
created with all the punctuation removed.

22
00:01:05,166 --> 00:01:10,166
So let's go back to the console
and let's press the up arrow to get

23
00:01:10,300 --> 00:01:14,000
the line of code that is giving us access
to the first review.

24
00:01:14,200 --> 00:01:16,900
Here it is. So let's press enter now.

25
00:01:16,900 --> 00:01:20,400
And as you can see, the thrilled
that's disappeared.

26
00:01:20,833 --> 00:01:24,600
Therefore that's exactly what the remove
punctuation function does here.

27
00:01:24,800 --> 00:01:28,666
It removes any kind of punctuation,
including dot commas.

28
00:01:28,666 --> 00:01:29,100
Collins.

29
00:01:29,100 --> 00:01:32,100
Semicolon
or any other kind of punctuation.

30
00:01:32,366 --> 00:01:33,100
All right.

31
00:01:33,100 --> 00:01:34,266
So next step done.

32
00:01:34,266 --> 00:01:36,433
We are ready to move on to the next step,

33
00:01:36,433 --> 00:01:40,066
which will be to remove
all the non relevant words in the reviews.

34
00:01:40,266 --> 00:01:43,300
So for example if we have a look at this
first review here.

35
00:01:43,300 --> 00:01:45,566
Well this is not very relevant.

36
00:01:45,566 --> 00:01:49,200
You know this doesn't give any hint on
knowing whether the review is positive

37
00:01:49,200 --> 00:01:50,100
or negative.

38
00:01:50,100 --> 00:01:53,833
So this is typically a word that we don't
want to have in the final sparse matrix

39
00:01:54,033 --> 00:01:55,666
because this is not relevant.

40
00:01:55,666 --> 00:01:58,666
So we will remove it in the next step
of the cleaning process.

41
00:01:58,700 --> 00:02:01,633
And we'll do the same
for all the other words of the same kind.

42
00:02:01,633 --> 00:02:03,366
So let's do that in the next tutorial.

43
00:02:03,366 --> 00:02:05,133
And until then enjoy machine learning.
