1
00:00:02,000 --> 00:00:10,730
In this session, let's understand the different types of data we all know the two broad categorizations

2
00:00:10,940 --> 00:00:15,000
when it comes to data, qualitative data and quantitative data.

3
00:00:15,290 --> 00:00:22,110
But if you really see a further subdivision is there between qualitative and quantitative data.

4
00:00:22,850 --> 00:00:25,780
Let's see quantitative data.

5
00:00:26,390 --> 00:00:30,200
We have continuous and discrete motor quantitative.

6
00:00:31,070 --> 00:00:36,400
The key difference is continuous can be measured on a continuum or scale.

7
00:00:37,580 --> 00:00:40,270
You can subdivide this meaningfully.

8
00:00:40,610 --> 00:00:47,900
For example, if someone says 169 centimeters is their height, you can represent it as one point six

9
00:00:47,950 --> 00:00:48,710
millimeters also.

10
00:00:48,710 --> 00:00:48,970
Right.

11
00:00:49,960 --> 00:00:55,340
That kind of subdividing of data is not possible and discrete.

12
00:00:55,600 --> 00:00:58,090
Yeah, I'm primarily referring to count of data.

13
00:00:58,630 --> 00:01:05,280
Suppose in a class I can say there are 30 students, I cannot say thirty point one students or point

14
00:01:05,290 --> 00:01:05,890
three students.

15
00:01:05,890 --> 00:01:06,150
Right.

16
00:01:07,480 --> 00:01:10,060
Do you understand the difference between discrete and continuous?

17
00:01:11,120 --> 00:01:15,360
Right now, let's see qualitative data in qualitative.

18
00:01:15,830 --> 00:01:19,850
Let's start with binary binaries, two values are possible, right?

19
00:01:20,210 --> 00:01:21,090
Pass fail.

20
00:01:21,140 --> 00:01:21,620
Yes.

21
00:01:21,620 --> 00:01:22,850
No, OK.

22
00:01:24,880 --> 00:01:30,510
Apart from binary, you have nominal ordinal in qualitative data.

23
00:01:31,400 --> 00:01:38,850
Ordinal is about an ordered series, nominal, the inherent ranking or order is not there.

24
00:01:39,620 --> 00:01:41,960
Examples are gender, right?

25
00:01:42,230 --> 00:01:42,850
Race.

26
00:01:43,340 --> 00:01:47,310
You can't say that male is better than a female, right.

27
00:01:47,480 --> 00:01:54,530
So there is no inherent ranking or order, whereas in the case of ordinal, that order or ranking is

28
00:01:54,530 --> 00:01:56,780
possible when it comes to performance.

29
00:01:58,250 --> 00:02:05,150
Grade is better than grade B, grade BS, but better than grade see, so on and so forth, right.

30
00:02:05,630 --> 00:02:13,550
So you need to ascertain what are the different types of data that are there in the dataset that are

31
00:02:13,550 --> 00:02:16,910
taking for analysis why this is important?

32
00:02:17,750 --> 00:02:23,980
You have to understand that algorithms are developed only for quantitative data.

33
00:02:24,620 --> 00:02:29,780
So the quantitative data must be converted into quantitative data.

34
00:02:29,990 --> 00:02:35,280
That is why we need to isolate the qualitative data in our dataset.

35
00:02:35,660 --> 00:02:37,770
So please keep this up in your mind.

36
00:02:38,300 --> 00:02:41,480
OK, now let's see some real life examples.

37
00:02:43,520 --> 00:02:50,340
In the case of insurance, we do this as part of deponent, an independent variable, right?

38
00:02:51,290 --> 00:02:52,850
I'm taking the same example.

39
00:02:52,950 --> 00:02:59,510
If you see this, the insurance charges that must be charged by the insurance company is dependent on

40
00:02:59,510 --> 00:03:02,240
so many factors that you really see.

41
00:03:02,750 --> 00:03:07,850
Why is numeric X1 is numeric, extra is numeric exploders.

42
00:03:07,850 --> 00:03:10,520
Normally the rest are non numeric.

43
00:03:11,440 --> 00:03:15,410
OK, you must convert them into numeric.

44
00:03:16,260 --> 00:03:16,630
Right.

45
00:03:17,250 --> 00:03:24,060
OK, the other example is the banking example in this case, if you see the wire itself is not numeric,

46
00:03:24,060 --> 00:03:24,420
right?

47
00:03:24,600 --> 00:03:27,180
It's a case of should loan be granted or not.

48
00:03:27,180 --> 00:03:35,040
Isabell, only number of dependents, applicants, income, koplik and income loan tenure and amount

49
00:03:35,040 --> 00:03:36,610
visa numeric data.

50
00:03:37,260 --> 00:03:39,810
The rest are all non numeric.

51
00:03:40,060 --> 00:03:42,350
OK, that is one, two, three, four, five.

52
00:03:42,360 --> 00:03:47,020
Only five out of 11 factors will have numeric data.

53
00:03:47,040 --> 00:03:48,990
The remaining will have no numeric.

54
00:03:50,460 --> 00:03:50,830
Right.

55
00:03:51,420 --> 00:03:56,790
So do you understand the different types of data that will be there in your data center?

56
00:03:57,890 --> 00:03:58,420
OK.