1
00:00:06,260 --> 00:00:09,750
Let us begin with the
fundamentals of visualization.

2
00:00:09,750 --> 00:00:13,175
What you see on the right
side is a picture graph or

3
00:00:13,175 --> 00:00:17,550
a bar chart prepared by
students in kindergarten class,

4
00:00:17,550 --> 00:00:19,200
four-year-olds at the St.

5
00:00:19,200 --> 00:00:21,240
Luke Catholic school
in Mississauga.

6
00:00:21,240 --> 00:00:23,130
This bar chart presents

7
00:00:23,130 --> 00:00:27,180
the frequency for
transportation modes

8
00:00:27,180 --> 00:00:28,810
the kids have used
to come to school.

9
00:00:28,810 --> 00:00:30,990
Some have been dropped
to school by car,

10
00:00:30,990 --> 00:00:34,110
others by bus, and two
students walked to school.

11
00:00:34,110 --> 00:00:37,465
In the brave, big
world of big data,

12
00:00:37,465 --> 00:00:39,950
we can see that children
are being trained at

13
00:00:39,950 --> 00:00:43,175
the earliest possible age
with data and data science.

14
00:00:43,175 --> 00:00:45,560
The most important
thing to realize is

15
00:00:45,560 --> 00:00:48,230
that the type of visualization

16
00:00:48,230 --> 00:00:50,720
you will use depends
upon the type of

17
00:00:50,720 --> 00:00:53,525
variables you're
trying to analyze.

18
00:00:53,525 --> 00:00:55,265
For instance, if
you're working with

19
00:00:55,265 --> 00:00:57,830
categorical variables
such as gender,

20
00:00:57,830 --> 00:00:59,840
you have to rely on

21
00:00:59,840 --> 00:01:02,900
a certain type of charting
tools than if you

22
00:01:02,900 --> 00:01:05,239
were to be working with
continuous variables

23
00:01:05,239 --> 00:01:08,200
such as age and income.

24
00:01:08,200 --> 00:01:10,750
In one case, you may
be using bar charts,

25
00:01:10,750 --> 00:01:12,155
in another case, you will be

26
00:01:12,155 --> 00:01:14,270
required to use scatter plots.

27
00:01:14,270 --> 00:01:17,180
I would like to draw
your attention to

28
00:01:17,180 --> 00:01:19,790
the Extreme Presentation Method

29
00:01:19,790 --> 00:01:22,835
developed by Dr. Andrew Abela.

30
00:01:22,835 --> 00:01:26,600
Essentially, it lays out the
possible ways of depicting

31
00:01:26,600 --> 00:01:28,370
data based on what kind of

32
00:01:28,370 --> 00:01:30,695
variables you have
at your disposal.

33
00:01:30,695 --> 00:01:32,765
This visualization, this graphic

34
00:01:32,765 --> 00:01:34,820
is developed by Dr. Abela that

35
00:01:34,820 --> 00:01:36,545
shows that if you
are interested in

36
00:01:36,545 --> 00:01:38,869
comparing variables
or demonstrating

37
00:01:38,869 --> 00:01:41,300
their distribution
or composition or

38
00:01:41,300 --> 00:01:44,365
the relationship between
two variables or more,

39
00:01:44,365 --> 00:01:46,790
you have to rely on
specific type of graph.

40
00:01:46,790 --> 00:01:50,150
If you're comparing items
with few categories,

41
00:01:50,150 --> 00:01:53,300
you can use bar charts
or column charts.

42
00:01:53,300 --> 00:01:56,690
If you are comparing behaviors
over time and if you

43
00:01:56,690 --> 00:02:01,040
have the time periods
running into several months,

44
00:02:01,040 --> 00:02:02,880
you may have to use a line chart.

45
00:02:02,880 --> 00:02:04,985
If the time periods
are not that many,

46
00:02:04,985 --> 00:02:08,080
then you can use columns
and other approaches.

47
00:02:08,080 --> 00:02:10,100
If you're trying to
depict relationships

48
00:02:10,100 --> 00:02:12,630
between two continuous variables,

49
00:02:12,630 --> 00:02:14,335
your choice is a scatterplot.

50
00:02:14,335 --> 00:02:18,110
The bubble chart will depict
two variables on x and

51
00:02:18,110 --> 00:02:20,720
y axis and the third variable

52
00:02:20,720 --> 00:02:22,610
will be depicted by the
size of the circle.

53
00:02:22,610 --> 00:02:24,680
So you can essentially
have three variables.

54
00:02:24,680 --> 00:02:26,060
If you're interested in depicting

55
00:02:26,060 --> 00:02:27,995
the distribution of the dataset,

56
00:02:27,995 --> 00:02:29,675
you can use the histogram,

57
00:02:29,675 --> 00:02:31,600
which could be a
bar chart type of

58
00:02:31,600 --> 00:02:34,060
histogram or a line histogram.

59
00:02:34,060 --> 00:02:35,540
The distribution could also be

60
00:02:35,540 --> 00:02:37,365
shown through scatter plots.

61
00:02:37,365 --> 00:02:39,105
If you would like to
show the composition,

62
00:02:39,105 --> 00:02:40,490
then if it's a static data,

63
00:02:40,490 --> 00:02:41,840
you can use pie charts.

64
00:02:41,840 --> 00:02:44,180
If you're showing
the composition that

65
00:02:44,180 --> 00:02:46,700
changes over time
for few periods,

66
00:02:46,700 --> 00:02:49,070
you can use stacked columns.

67
00:02:49,070 --> 00:02:51,065
If you have several periods,

68
00:02:51,065 --> 00:02:54,245
you can use the
stacked area charts.

69
00:02:54,245 --> 00:02:56,225
For hands-on in Python,

70
00:02:56,225 --> 00:02:58,610
we will be using the
Seaborn library and

71
00:02:58,610 --> 00:03:00,350
the Matplotlib library to

72
00:03:00,350 --> 00:03:02,755
create visualizations
in the labs.

73
00:03:02,755 --> 00:03:05,390
We will learn how to use
different functions within

74
00:03:05,390 --> 00:03:08,850
the library to create
different kinds of charts.