1
00:00:02,480 --> 00:00:05,900
So let's get started with the exploratory data analysis.

2
00:00:06,500 --> 00:00:09,470
The first step is understanding the data.

3
00:00:10,320 --> 00:00:17,670
I want to understand what is really there in my mind, and the first step for that is determining what

4
00:00:17,670 --> 00:00:20,520
is my independent variable and what is my independent.

5
00:00:21,900 --> 00:00:30,140
Let me explain this concept by taking a very simple example, a restaurant that provides pizza and home

6
00:00:30,150 --> 00:00:31,140
delivers pizza.

7
00:00:33,450 --> 00:00:40,920
That restaurant's time taken to deliver pizza is dependent on the order volume right now, many, many

8
00:00:40,920 --> 00:00:41,650
orders are there.

9
00:00:42,660 --> 00:00:44,950
The time taken to deliver the pizza.

10
00:00:45,480 --> 00:00:47,440
The time taken to prepare the pizza.

11
00:00:47,640 --> 00:00:50,390
Is cooking time and the oven temperature.

12
00:00:51,270 --> 00:00:55,890
So all the time taken to deliver is dependent on these four factors.

13
00:00:56,920 --> 00:01:04,180
So this is my dependent variable and these are the independent variables.

14
00:01:05,700 --> 00:01:06,010
Right.

15
00:01:06,720 --> 00:01:14,280
This is the first thing we acidy if I were to develop a machine learning model for this example, it

16
00:01:14,280 --> 00:01:14,730
would be.

17
00:01:16,300 --> 00:01:23,980
What will be the time taken to deliver pizza based on all our cooking time, oven temperature and delivery?

18
00:01:25,090 --> 00:01:25,290
Right.

19
00:01:26,170 --> 00:01:29,410
So you get an idea of what is an independent variable and dependent variable.

20
00:01:30,930 --> 00:01:32,340
Another example would be.

21
00:01:34,000 --> 00:01:37,690
The member of Marks is caught in an exam, he's dependent.

22
00:01:38,950 --> 00:01:41,620
On the number of this study.

23
00:01:42,960 --> 00:01:52,020
Right, so the moxa pain is the dependent variable and the number of hours I study is the independent.

24
00:01:53,600 --> 00:01:53,910
Right.

25
00:01:55,090 --> 00:02:01,110
So let's see the scenario that we will be taking in our exploratory data analysis.

26
00:02:02,050 --> 00:02:09,280
Will be taking an insurance he studied, the insurance company has to determine the insurance charges

27
00:02:09,280 --> 00:02:11,980
for an individual who has taken the whole home.

28
00:02:14,110 --> 00:02:21,520
The insurance company will evaluate leverne parameters and determine the insurance charges for the individual,

29
00:02:22,330 --> 00:02:28,300
as you can see, the parameters of gender married, whether the individual is married or not, a member

30
00:02:28,300 --> 00:02:34,660
of the police education level, whether the individual is self-employed, what is applicant's income?

31
00:02:35,500 --> 00:02:37,530
What is the applicant's income?

32
00:02:37,570 --> 00:02:38,820
What is the loan amount?

33
00:02:38,830 --> 00:02:42,130
What is the training or what is a credit history?

34
00:02:43,000 --> 00:02:48,610
If the individual has taken loans in the past, what is the property?

35
00:02:49,000 --> 00:02:51,790
So all these things are ascertained by the.

36
00:02:54,090 --> 00:03:00,210
Insurance companies, as you can see, that I own my summers and American summers normally, right?

37
00:03:02,870 --> 00:03:08,300
And if you see here, the insurance charges is dependent on these factors, health insurance charges

38
00:03:08,300 --> 00:03:13,990
is the dependent variable and all the extras that you see here are independent votes.

39
00:03:15,660 --> 00:03:16,140
Correct.

40
00:03:17,070 --> 00:03:23,480
So this is the fastest way we are going to be using this example through the exploratory data analysis.

41
00:03:26,110 --> 00:03:30,840
After this, you want to understand what is actually there in the data, right?

42
00:03:31,810 --> 00:03:38,710
For that, you will use data not ahead of time, it tells you the top three records that is there in

43
00:03:38,710 --> 00:03:41,210
my data can got an idea what is there?

44
00:03:41,550 --> 00:03:44,770
You can see that is also a case of.

45
00:03:45,910 --> 00:03:47,720
Probably a junk or missing one.

46
00:03:48,370 --> 00:03:49,760
How are you going to treat that?

47
00:03:50,260 --> 00:03:54,670
That is one of the things that we are going to cover in these subsequent sections.

48
00:03:55,130 --> 00:04:01,390
Remember, if you don't address a lot of junk values, your forecast accuracy will take a hit and probably

49
00:04:01,390 --> 00:04:03,400
you may even get an error like.

50
00:04:05,760 --> 00:04:11,640
Point out we'll be using Google Cola and Python, we will collaborate the development, environment

51
00:04:11,640 --> 00:04:13,500
and Python as the programming language.

52
00:04:14,250 --> 00:04:21,240
So the fight that I will be using for analysis file containing the past data is actually stored in the

53
00:04:21,240 --> 00:04:22,210
Google brain.

54
00:04:22,860 --> 00:04:26,960
I first need to establish the connection between CoLab and Google.

55
00:04:27,300 --> 00:04:27,520
Right.

56
00:04:28,020 --> 00:04:29,190
And after that.

57
00:04:30,110 --> 00:04:35,630
I import five pounds and then I remember this is my prime location.

58
00:04:36,550 --> 00:04:38,170
The Firestone Google.

59
00:04:39,250 --> 00:04:39,560
Yeah.

60
00:04:42,190 --> 00:04:45,550
Damn, I want to understand what type of data is there in my dataset.

61
00:04:46,530 --> 00:04:49,080
Like, you see, these are the variables that you saw.

62
00:04:50,460 --> 00:04:50,780
Right.

63
00:04:52,360 --> 00:04:58,080
If I use data, not info, I get, whether it is a flawed or an integer or an object.

64
00:04:58,660 --> 00:05:03,550
I get all this type of information.

65
00:05:04,660 --> 00:05:05,020
Right.

66
00:05:05,440 --> 00:05:12,070
So this is part of understanding that it is data mining dependent versus independent variables.

67
00:05:14,800 --> 00:05:23,890
Understanding what variables are, what variables anonymously play into the float, was stress getting

68
00:05:24,310 --> 00:05:30,850
an idea about the type of data, the kind of data that we have by looking at the top three records?

69
00:05:32,230 --> 00:05:36,050
So that's what you do in understanding it, right?

70
00:05:36,460 --> 00:05:39,740
So the next step is doing univariate analysis.

71
00:05:40,650 --> 00:05:41,050
OK.