1
00:00:00,000 --> 00:00:05,073
Now before we start using Gephi, let me
tell you about the software tools that we

2
00:00:05,073 --> 00:00:10,019
will and won't be using.
In this class, we'll be using Gephi for

3
00:00:10,019 --> 00:00:14,023
visualization and for calculating simple
network metrics.

4
00:00:14,023 --> 00:00:20,018
We'll be using NetLogo to understand the
dynamics that shape network formation and

5
00:00:20,018 --> 00:00:23,079
also to understand processes occurring on
networks.

6
00:00:23,079 --> 00:00:29,003
And finally for those of you doing
programming assignments we'll be using

7
00:00:29,003 --> 00:00:33,007
iGraph for more sophisticated, more
flexible calculations.

8
00:00:33,028 --> 00:00:39,016
There are actually a lot of alternatives,
and let me tell you about some of them,

9
00:00:39,016 --> 00:00:45,011
because once you know what capabilities
you like, you can readily find them in

10
00:00:45,011 --> 00:00:49,082
other software that may be more
well-suited to your projects.

11
00:00:49,082 --> 00:00:54,687
So there's the software package called
Pajek which means spider in Slovenian.

12
00:00:54,687 --> 00:00:59,482
It has been developed for a number of
years, it's a very mature software

13
00:00:59,482 --> 00:01:02,779
package.
It has very extensive functionality which

14
00:01:02,779 --> 00:01:08,099
you access through drop down menus.
And the reason why we're not using it even

15
00:01:08,099 --> 00:01:13,544
though I have used it when teaching this
type of class in the past is because it

16
00:01:13,544 --> 00:01:18,113
only works on the Windows platform and
some students have had trouble using

17
00:01:18,113 --> 00:01:22,877
Windows emulators on other systems.
So I just made the choice to switch over

18
00:01:22,877 --> 00:01:28,017
to platform-independent software.
However Pajek might be a very good choice.

19
00:01:28,017 --> 00:01:33,001
In addition, there is a book, a textbook,
called Social Network and, Exploratory

20
00:01:33,001 --> 00:01:38,003
Social Network Analysis with Pajek, which
guides you through many analyses that you

21
00:01:38,003 --> 00:01:42,069
can do with that software.
We also will not be using UCINet, which

22
00:01:42,069 --> 00:01:48,056
again is fairly mature, or very mature.
It has extensive functionality, mostly

23
00:01:48,056 --> 00:01:54,097
focused around sociology so, I think until
recently UCI Net did not really scale up

24
00:01:54,097 --> 00:01:58,668
to large networks.
Unfortunately it's, it's also Windows

25
00:01:58,668 --> 00:02:05,402
only, and the license may cost you a
couple hundred dollars if you're not a, a,

26
00:02:05,402 --> 00:02:10,018
a full time student, or somehow affiliated
with the university.

27
00:02:10,018 --> 00:02:14,039
We also will be using NodeXL, NodeXL is a
relative newcomer.

28
00:02:14,039 --> 00:02:18,047
Its social network analysis neatly
integrated into Excel.

29
00:02:18,068 --> 00:02:22,083
It's still, I think, very much in beta,
it's being developed.

30
00:02:22,083 --> 00:02:27,019
However, it's free.
And if you're, are already an Excel user,

31
00:02:27,019 --> 00:02:31,090
then this may be a very friendly way to do
social network analysis.

32
00:02:31,090 --> 00:02:37,075
Again, it's, it's actually Windows only,
only, the, Windows version of Excel

33
00:02:37,075 --> 00:02:41,081
supports NodeXL, and that's why we're not
using it.

34
00:02:41,081 --> 00:02:47,030
If you're a programmer, there's an
excellent alternative called NetworkX.

35
00:02:47,030 --> 00:02:53,025
It has extensive functionality and it's
actually built upon, Fortran and C

36
00:02:53,025 --> 00:02:58,067
libraries, which are optimized to scale
for very, very large matrices.

37
00:02:58,067 --> 00:03:02,081
And so you can do calculations for very
large networks.

38
00:03:02,081 --> 00:03:09,233
It's, on top of that it's open source, so
that's definitely something to consider if

39
00:03:09,233 --> 00:03:14,060
you have this background.
There are many, many other alternatives,

40
00:03:14,060 --> 00:03:20,417
most of them are in various phases of
development one very mature package is the

41
00:03:20,417 --> 00:03:27,035
social network analysis package for R, it,
it has a kind of statistics heavy duty

42
00:03:27,035 --> 00:03:33,033
functionality for example, p-star models
which we won't even get into in, in this

43
00:03:33,033 --> 00:03:37,094
course, are supported in this social
network analysis package.

44
00:03:38,015 --> 00:03:43,092
In addition, if you are interested some,
p-star models are used to model networks

45
00:03:43,092 --> 00:03:48,057
as they evolve to time.
Just a hypothesis about, what's driving

46
00:03:48,057 --> 00:03:52,049
edge formation.
And if you're interested in visualizing

47
00:03:52,049 --> 00:03:57,070
networks over time with kind of an
integrated P star model functionality,

48
00:03:57,070 --> 00:04:02,077
there's the package SoNIA, that, that you
can use for this purpose.

49
00:04:02,077 --> 00:04:08,507
Now let's get to our main task for this
lecture, which is to work with Gephi.

50
00:04:08,507 --> 00:04:12,513
So what I'd like you to do is to download
Gephi.

51
00:04:12,513 --> 00:04:17,165
It's free, it should work on any platform,
and install it.

52
00:04:17,165 --> 00:04:23,453
Once you've installed it, download the
data file dining.gephi from the Coursera

53
00:04:23,453 --> 00:04:27,490
folder for this lecture, and we can start
to play.

54
00:04:27,490 --> 00:04:33,096
Now that you downloaded Gephi and
installed it, we can start to play with

55
00:04:33,096 --> 00:04:38,531
this dining.gephi data file.
Now I've opened it recently so it's shown

56
00:04:38,531 --> 00:04:45,585
in this helpful welcome screen but perhaps
you don't have that so you can go to File,

57
00:04:45,585 --> 00:04:52,224
Open, and navigate to the directory where
you've saved it and then just double click

58
00:04:52,224 --> 00:04:56,221
to open it.
And here we're waiting for it to load and

59
00:04:56,221 --> 00:05:03,016
pretty soon it's going to tell us, for
example, the number of nodes and edges.

60
00:05:03,016 --> 00:05:07,005
It's going to tell us there are 26 nodes
and 52 edges.

61
00:05:07,005 --> 00:05:11,080
What kind of data is this?
Well, this was data gathered in a girls'

62
00:05:11,080 --> 00:05:17,062
dormitory back into the 1940s, where the
girls were asked to name their first and

63
00:05:17,062 --> 00:05:22,001
second choices for who they would like to
sit with at dinner.

64
00:05:22,001 --> 00:05:25,039
So, this is the dining table partners'
data set.

65
00:05:26,024 --> 00:05:32,012
You, you can see a random layout here, and
I've done this on purpose to demonstrate

66
00:05:32,012 --> 00:05:38,000
how even with such a small network as this
one, a random layout can really obscure

67
00:05:38,000 --> 00:05:43,031
whatever insights you're trying to see.
Now, Gephi helps a little bit.

68
00:05:43,031 --> 00:05:48,677
Here we can tell for example, that this
girl is relatively popular because lots of

69
00:05:48,677 --> 00:05:53,063
other girls have named her as their first
or second choice.

70
00:05:53,063 --> 00:05:59,030
Now, if we want to know which girl exactly
this is, what we do is we select this

71
00:05:59,030 --> 00:06:01,093
little arrow.
Plus question mark.

72
00:06:01,093 --> 00:06:09,027
It's actually an, brings up an edit window
over here, and now when we left-click on

73
00:06:09,027 --> 00:06:15,541
this node, we find out that this girl's
name is Eva, in addition to some other

74
00:06:15,541 --> 00:06:20,081
properties.
We also may want to know the overall

75
00:06:20,081 --> 00:06:26,096
structure so we can select a layout.
And here I went to the layout panel, and

76
00:06:26,096 --> 00:06:32,045
I'm going to choose a layout.
Force Atlas two is one of my favorites,

77
00:06:32,045 --> 00:06:38,043
but Yen Phen Hu is also very fine.
It turns out it doesn't matter too much

78
00:06:38,043 --> 00:06:44,042
when the network is this small.
It's more with large networks that you run

79
00:06:44,042 --> 00:06:49,018
into utilization difficulties.
So, let's try Force Atlas two.

80
00:06:49,018 --> 00:06:54,049
And I'm going to click on run.
There're all of these different parameters

81
00:06:54,049 --> 00:07:00,037
which you can tweak and, and play with
that have to do with how the model's

82
00:07:00,037 --> 00:07:04,000
running.
Essentially, all of these different layout

83
00:07:04,000 --> 00:07:09,081
algorithms want to place nodes that are
linked close together and ones that are

84
00:07:09,081 --> 00:07:12,094
not.
It wants them to repel from each other.

85
00:07:12,094 --> 00:07:19,016
So we're going to say run.
And it ran, it's doing a little bit of

86
00:07:19,016 --> 00:07:23,099
spinning.
Let's see if it'll actually spin into a

87
00:07:23,099 --> 00:07:29,060
good position.
Oh, indeed, very nice.

88
00:07:29,087 --> 00:07:32,026
Okay.
Let's stop it right there.

89
00:07:32,026 --> 00:07:38,151
And you may notice that now some nodes are
out of use, so you can click this little

90
00:07:38,151 --> 00:07:44,184
magnifying glass that will center it on
the graph but it may end up looking a

91
00:07:44,184 --> 00:07:49,041
little bit too small.
So I'm going to use the track pad or you

92
00:07:49,041 --> 00:07:55,364
can use your mouse wheel to just zoom in.
And then I'm going to right click and drag

93
00:07:55,364 --> 00:08:00,075
and zoom in some more.
And this I think shows off the layout of

94
00:08:00,075 --> 00:08:03,720
the girls' preferences a little bit more
clearly.

95
00:08:03,720 --> 00:08:07,441
You may want to tweak things just a little
bit.

96
00:08:07,441 --> 00:08:12,055
I'm going to select the hand to drag the
node just here.

97
00:08:12,055 --> 00:08:17,529
It wasn't quite clear who was, who was
picking whom, so maybe I just move it a

98
00:08:17,529 --> 00:08:22,774
little bit, right, so a, a little bit of
minor adjustment is no big deal.

99
00:08:22,774 --> 00:08:29,070
If you're moving things around because you
want them to look a certain way to prove

100
00:08:29,070 --> 00:08:32,416
some point, then I wouldn't, I wouldn't do
this.

101
00:08:32,416 --> 00:08:37,931
But just for us and wanting to see who
exactly has picked whom, I think it's

102
00:08:37,931 --> 00:08:39,513
perfectly fine.
Okay.

103
00:08:39,513 --> 00:08:46,154
Some other things that you can do is you
can change the color for all the nodes.

104
00:08:46,154 --> 00:08:52,237
So, if we want them all to be red, and
then we, and so I right-click to choose

105
00:08:52,237 --> 00:08:57,000
the color, and then I left-click to set
them all to red.

106
00:08:57,023 --> 00:09:01,038
I may want to, yes okay, these were a
little overlapped.

107
00:09:01,038 --> 00:09:07,024
So we just want to avoid overlaps where we
don't know where there's an edge.

108
00:09:07,024 --> 00:09:13,726
Is incident on the node, or the node is
over the edge and it's actually obscuring

109
00:09:13,726 --> 00:09:17,064
it.
We can also adjust the size if they seem a

110
00:09:17,064 --> 00:09:21,072
little bit big.
I'm going to right click on the size.

111
00:09:21,072 --> 00:09:25,064
Well actually, 1.0, that's already quite,
quite fine.

112
00:09:25,064 --> 00:09:29,017
I guess you could try a fraction of the
size.

113
00:09:29,017 --> 00:09:34,058
I'm going to left-click and now they're
tiny, so let's go back to 1.0.

114
00:09:35,094 --> 00:09:40,079
Inset that.
Oops, alright.

115
00:09:40,079 --> 00:09:42,041
Let's see.
Five?

116
00:09:42,041 --> 00:09:46,050
Apparently it wasn't five before.
Okay.

117
00:09:46,050 --> 00:09:50,026
There we go.
So they're a size five.

118
00:09:50,059 --> 00:09:57,051
We may find this more informative if
instead of having to click on this edit

119
00:09:57,051 --> 00:10:04,000
attribute where we then click on
individual nodes and find out, hey, this

120
00:10:04,000 --> 00:10:08,009
girl is Hazel.
And this one is Maxine, this is where I'm

121
00:10:08,009 --> 00:10:12,000
seeing all this info after I've selected
the Edit tool.

122
00:10:12,000 --> 00:10:18,006
Perhaps if we just show the labels for all
the nodes, so I'm just gonna click on this

123
00:10:18,006 --> 00:10:23,430
T down here that says show node labels.
And they are way too big so we're just

124
00:10:23,430 --> 00:10:27,514
going to have to try and scale then down a
little bit.

125
00:10:27,514 --> 00:10:32,080
There we go.
And, we can go back to the layout and say

126
00:10:32,080 --> 00:10:35,517
label adjust.
Try to move these around so that the

127
00:10:35,517 --> 00:10:39,336
labels don't overlap as much, and it's
actually done it.

128
00:10:39,336 --> 00:10:43,647
I don't know if you could see it, it
popped up very quickly.

129
00:10:43,647 --> 00:10:46,659
And now, all the labels are fairly
visible.

130
00:10:46,659 --> 00:10:53,074
Another thing that we may want to do is to
color the edges differently depending on

131
00:10:53,074 --> 00:10:56,688
whether this is a girl's first or second
choice.

132
00:10:56,688 --> 00:11:00,962
So I'm going to go to the partition tab,
click on it.

133
00:11:00,962 --> 00:11:06,473
I'm going to click on edges and I'm going
to click this little reload.

134
00:11:06,473 --> 00:11:12,878
Load the different partitions or actually
characteristics of the edges that are

135
00:11:12,878 --> 00:11:16,796
discreet categories and I'm going to click
on label.

136
00:11:16,796 --> 00:11:22,673
And this has whether it was first choice
or second choice, so I am going to click

137
00:11:22,673 --> 00:11:26,195
apply.
And now we see all of the second choices

138
00:11:26,195 --> 00:11:31,943
in blue and the first choices in red.
And I will start here because we need to

139
00:11:31,943 --> 00:11:37,512
explain some additional concepts before we
can use Gephi, oh, but actually, let me

140
00:11:37,512 --> 00:11:40,906
show you.
So, we're we are right now in, in overview

141
00:11:40,906 --> 00:11:45,097
mode where we have all these different
tools that we can use.

142
00:11:45,097 --> 00:11:53,032
We can also click on data laboratory, And
in data laboratory, we have a spreadsheet

143
00:11:53,032 --> 00:12:00,049
of all of the node information, and if we
click on edges, we have all of the edge

144
00:12:00,049 --> 00:12:06,003
information as well.
And finally in Preview we can preview the

145
00:12:06,003 --> 00:12:13,082
visualization that then we can export.
So I'm going to click Refresh now to see

146
00:12:13,082 --> 00:12:17,708
it.
I can, I should probably see the node

147
00:12:17,708 --> 00:12:22,076
labels so I'm going to click that and hit
refresh.

148
00:12:22,076 --> 00:12:29,076
And then, I'm going to unclick curve so
that we get the arrows for the

149
00:12:29,076 --> 00:12:37,099
directionality of the edge and perhaps I
can also well change the background, in

150
00:12:37,099 --> 00:12:44,012
this case maybe I want a blue background
or something like that.

151
00:12:44,012 --> 00:12:48,018
No, not a blue background.
Let's go back to white.

152
00:12:48,018 --> 00:12:54,077
And then if I wanted to export this I
would click this button down here and I

153
00:12:54,077 --> 00:13:00,042
could choose whether to export it as a pdf
or a png or an sv2 file.

154
00:13:00,042 --> 00:13:04,039
I'm not going to do that right now because
we have some more work to do.

155
00:13:04,039 --> 00:13:10,081
So I'm going to go back to overview and
stop here while we learn a few more

156
00:13:10,081 --> 00:13:13,052
concepts and then we'll be back.