1
00:00:01,770 --> 00:00:05,700
The next thing we are going to do is known as variable transformation.

2
00:00:07,260 --> 00:00:12,420
If you don't notice that we have four different distance variables to represent the same information.

3
00:00:12,690 --> 00:00:20,130
That is, we just want to get the effect of having employment hubs nearby this town.

4
00:00:20,760 --> 00:00:24,420
And for that, we had having four different variables.

5
00:00:25,770 --> 00:00:27,940
This is already presenting a single feature.

6
00:00:29,970 --> 00:00:34,140
So we'll be transforming these four different variables into one single variable.

7
00:00:34,830 --> 00:00:36,360
We can do it in multiple ways.

8
00:00:36,780 --> 00:00:43,410
One of the ways is to use the maximum of all these four or we can to take the minimum of all this forward,

9
00:00:44,370 --> 00:00:46,200
or we can take the average of these four.

10
00:00:47,940 --> 00:00:50,790
Usually we have to do whatever makes more business sense.

11
00:00:51,780 --> 00:00:58,660
Here we will be taking the average of these four distances to get one average distance value, which

12
00:00:58,660 --> 00:01:03,910
we will use to represent the feature of employment activity near that particular town.

13
00:01:05,490 --> 00:01:10,120
So to create that variable, we will right be if dollar average distance.

14
00:01:15,490 --> 00:01:16,470
This video will get.

15
00:01:19,060 --> 00:01:21,140
Some of those four divided by four.

16
00:01:21,730 --> 00:01:26,460
So the taller dollar, this one plus the M dollar.

17
00:01:26,510 --> 00:01:27,230
This to.

18
00:01:32,090 --> 00:01:32,670
Misty.

19
00:01:36,970 --> 00:01:37,660
And just for.

20
00:01:42,910 --> 00:01:43,970
Divided by four.

21
00:01:46,230 --> 00:01:47,460
Let's create this variable.

22
00:01:48,180 --> 00:01:51,100
Now we'll click on DBI, a variable the does it.

23
00:01:52,440 --> 00:01:55,490
You can see in the end.

24
00:01:55,560 --> 00:01:56,820
We have another column.

25
00:01:57,660 --> 00:02:03,870
You can see that the four point zero eight value is actually the average of those forward values in

26
00:02:03,870 --> 00:02:04,090
this.

27
00:02:04,090 --> 00:02:05,850
To add this to this three individual.

28
00:02:08,630 --> 00:02:14,690
Now that we have this video being created, we have to delete the other four based variables from our

29
00:02:14,690 --> 00:02:15,170
dataset.

30
00:02:17,540 --> 00:02:25,160
To delete the four variables, we need to get that position in the dataset, so we need to find out

31
00:02:25,160 --> 00:02:31,460
the location of Difford distance variables that we want to remove to find out the location of the variable.

32
00:02:32,840 --> 00:02:38,810
If you bought over the name of that column, you can see that this is column number six.

33
00:02:40,280 --> 00:02:42,290
This is column seven, eight, nine.

34
00:02:42,620 --> 00:02:45,830
We want to remove columns six to nine.

35
00:02:47,660 --> 00:02:53,930
While we are removing a column, we can create a new dataset just to ensure that when we delete the

36
00:02:53,930 --> 00:02:58,910
variables, we do not accidentally delete these variables that we do not want to delete.

37
00:03:00,140 --> 00:03:03,740
So let us create a new dataset once, which is D.F. to.

38
00:03:08,640 --> 00:03:11,210
It will get all relisten, be it.

39
00:03:11,490 --> 00:03:12,530
So it is D.F..

40
00:03:13,380 --> 00:03:14,760
We'll start with square brackets.

41
00:03:15,540 --> 00:03:17,790
The first parameter is for the rows.

42
00:03:18,030 --> 00:03:20,700
We want all the rows, so we will not mention anything.

43
00:03:20,830 --> 00:03:22,080
We will straightaway put a comma.

44
00:03:22,590 --> 00:03:24,060
The second bedroom I that is columns.

45
00:03:24,090 --> 00:03:28,450
We want to include we do not want to include column six to nine.

46
00:03:28,560 --> 00:03:29,910
So we'll use a minus sign.

47
00:03:30,590 --> 00:03:33,150
So it is minus six to minus nine.

48
00:03:35,330 --> 00:03:36,030
Let's run this.

49
00:03:36,560 --> 00:03:38,000
We could be off to No.

50
00:03:38,730 --> 00:03:39,860
Just click on, be up to.

51
00:03:43,880 --> 00:03:49,540
You can see that this one, this two, this three, just for all these four variables are now deleted

52
00:03:49,880 --> 00:03:55,070
since we deleted the correct variables and B of two is the dataset, which we want to use.

53
00:03:55,400 --> 00:04:01,340
We will assign deserve to to be if one that is D.F., so be F is equal to be of two.

54
00:04:07,470 --> 00:04:11,100
Now you can see that B.F. has 15 variables instead of 20.

55
00:04:12,660 --> 00:04:18,330
We can remove B of two by writing item and within records we will be off to.

56
00:04:22,940 --> 00:04:24,420
So B of two is removed.

57
00:04:25,500 --> 00:04:31,020
And we have the D variable, which has 16 variables.

58
00:04:31,410 --> 00:04:37,410
Now, if you remember the three points that we know there, one point was to remove the bus terminal

59
00:04:37,410 --> 00:04:37,890
where even.

60
00:04:40,500 --> 00:04:44,500
To remove that variable, we will again go to the data set.

61
00:04:48,490 --> 00:04:57,100
And find out the location of bus terminal variable, if I hold already hitting it tells me that it is

62
00:04:57,100 --> 00:04:58,090
column number 13.

63
00:04:58,720 --> 00:05:03,880
So I will go and read B.F. Gate.

64
00:05:07,260 --> 00:05:12,300
D.F. Square Records, Gomaa, minus 30.

65
00:05:14,840 --> 00:05:26,900
And this you can see now we have 15 variables just to confirm the flick again and check if I'm in a

66
00:05:26,900 --> 00:05:28,490
little mood and it is.

67
00:05:30,790 --> 00:05:37,100
So this is how we removed the variables that we do not want in a dataset.

68
00:05:40,300 --> 00:05:47,480
I go back to the three points that we noted in the summary of edict of univariate analysis.

69
00:05:52,560 --> 00:05:59,570
So we have handled the outliers by capping them in rainfall and and hard rooms.

70
00:06:01,370 --> 00:06:07,100
We have handled the missing values by replacing them with the mean values in an ice bed.

71
00:06:08,150 --> 00:06:10,520
And we have diluted this variable called Busted.

72
00:06:11,330 --> 00:06:15,620
And we have also transformed before distance what he would do.

73
00:06:15,860 --> 00:06:23,210
One common average distance variable and we'll remove those other for distance valuables in the coming

74
00:06:23,210 --> 00:06:23,620
videos.

75
00:06:23,660 --> 00:06:26,420
We will learn how to handle categorical variables.
