pacman::p_load(igraph, tidygraph, ggraph,
visNetwork, lubridate, clock,
tidyverse, graphlayouts)Hands-on Exercise 6 - Modelling, Visualising and Analysing Network Data with R
Learning Objectives:
create graph object data frames, manipulate them using appropriate functions of dplyr, lubridate, and tidygraph,
build network graph visualisation using appropriate functions of ggraph,
compute network geometrics using tidygraph,
build advanced graph visualisation by incorporating the network geometrics, and
build interactive network visualisation using visNetwork package.
Getting Started
Installing and loading the required libraries
The following R packages will be used:
igraph
tidygraph
ggraph
visNetwork
tidyverse
Code chunk below will be used to check if these packages have been installed and also will load them into the working R environment.
The Data
The data sets used are from an oil exploration and extraction company. There are two data sets:
Nodes data: GAStech-email_edges.csv which consists of two weeks of 9063 emails correspondances between 55 employees.
edges (aka link) data: GAStech_email_nodes.csv which consist of the names, department and title of the 55 employees.
Importing network data from files
Import GAStech_email_node.csv and GAStech_email_edges-v2.csv into RStudio environment by using read_csv() of readr package.
GAStech_nodes <- read_csv("data/GAStech_email_node.csv")
GAStech_edges <- read_csv("data/GAStech_email_edge-v2.csv")Reviewing the imported data
glimpse(GAStech_edges)Rows: 9,063
Columns: 8
$ source <dbl> 43, 43, 44, 44, 44, 44, 44, 44, 44, 44, 44, 44, 26, 26, 26…
$ target <dbl> 41, 40, 51, 52, 53, 45, 44, 46, 48, 49, 47, 54, 27, 28, 29…
$ SentDate <chr> "6/1/2014", "6/1/2014", "6/1/2014", "6/1/2014", "6/1/2014"…
$ SentTime <time> 08:39:00, 08:39:00, 08:58:00, 08:58:00, 08:58:00, 08:58:0…
$ Subject <chr> "GT-SeismicProcessorPro Bug Report", "GT-SeismicProcessorP…
$ MainSubject <chr> "Work related", "Work related", "Work related", "Work rela…
$ sourceLabel <chr> "Sven.Flecha", "Sven.Flecha", "Kanon.Herrero", "Kanon.Herr…
$ targetLabel <chr> "Isak.Baza", "Lucas.Alcazar", "Felix.Resumir", "Hideki.Coc…Wrangling time
The output report of GAStech_edges below reveals that the SentDate is treated as “Character” data type instead of date data type, which is incorrect. Thus, the data type of SentDate field needs to be changed to “Date”” data type.
Note:
both dmy() and wday() are functions of lubridate package. lubridate is an R package that makes it easier to work with dates and times.
dmy() transforms the SentDate to Date data type.
wday() returns the day of the week as a decimal number or an ordered factor if label is TRUE. The argument abbr is FALSE keep the daya spells in full, i.e. Monday. The function will create a new column in the data.frame i.e. Weekday and the output of wday() will save in this newly created field.
the values in the Weekday field are in ordinal scale.
GAStech_edges <- GAStech_edges %>%
mutate(SendDate = dmy(SentDate)) %>%
mutate(Weekday = wday(SentDate,
label = TRUE,
abbr = FALSE))Reviewing the revised date fields
glimpse(GAStech_edges)Rows: 9,063
Columns: 10
$ source <dbl> 43, 43, 44, 44, 44, 44, 44, 44, 44, 44, 44, 44, 26, 26, 26…
$ target <dbl> 41, 40, 51, 52, 53, 45, 44, 46, 48, 49, 47, 54, 27, 28, 29…
$ SentDate <chr> "6/1/2014", "6/1/2014", "6/1/2014", "6/1/2014", "6/1/2014"…
$ SentTime <time> 08:39:00, 08:39:00, 08:58:00, 08:58:00, 08:58:00, 08:58:0…
$ Subject <chr> "GT-SeismicProcessorPro Bug Report", "GT-SeismicProcessorP…
$ MainSubject <chr> "Work related", "Work related", "Work related", "Work rela…
$ sourceLabel <chr> "Sven.Flecha", "Sven.Flecha", "Kanon.Herrero", "Kanon.Herr…
$ targetLabel <chr> "Isak.Baza", "Lucas.Alcazar", "Felix.Resumir", "Hideki.Coc…
$ SendDate <date> 2014-01-06, 2014-01-06, 2014-01-06, 2014-01-06, 2014-01-0…
$ Weekday <ord> Friday, Friday, Friday, Friday, Friday, Friday, Friday, Fr…Wrangling attributes
A close examination of GAStech_edges data.frame reveals that it consists of individual e-mail flow records, which is not very useful for visualisation.
Thus, aggregation is done for the individual by date, senders, receivers, main subject and day of the week.
four functions from dplyr package are used. They are: filter(), group(), summarise(), and ungroup().
The output data.frame is called GAStech_edges_aggregated.
A new field called Weight has been added in GAStech_edges_aggregated.
GAStech_edges_aggregated <- GAStech_edges %>%
filter(MainSubject == "Work related") %>%
group_by(source, target, Weekday) %>%
summarise(Weight = n()) %>%
filter(source!=target) %>%
filter(Weight > 1) %>%
ungroup()Reviewing the revised edges file
glimpse(GAStech_edges_aggregated)Rows: 1,372
Columns: 4
$ source <dbl> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,…
$ target <dbl> 2, 2, 2, 2, 2, 3, 3, 3, 3, 3, 4, 4, 4, 4, 4, 5, 5, 5, 5, 5, 6,…
$ Weekday <ord> Sunday, Monday, Tuesday, Wednesday, Friday, Sunday, Monday, Tu…
$ Weight <int> 5, 2, 3, 4, 6, 5, 2, 3, 4, 6, 5, 2, 3, 4, 6, 5, 2, 3, 4, 6, 5,…Creating network objects using tidygraph
tidygraph package provides a tidy API for graph/network manipulation. Network data can be envisioned as 2 tidy tables, 1 for node data and 1 for edge data.
tidygraph provides a way to switch between the two tables and provides dplyr verbs for manipulating them. It is also provides access to many graph algorithms with return values that facilitate their use in a tidy workflow.
The tbl_graph object
tidygraph can be used to create network objects:
tbl_graph()creates a tbl_graph network object from nodes and edges data.as_tbl_graph()converts network data and objects to a tbl_graph network. Below are network data and objects supported byas_tbl_graph()a node data.frame and an edge data.frame,
data.frame, list, matrix from base,
igraph from igraph,
network from network,
dendrogram and hclust from stats,
Node from data.tree,
phylo and evonet from ape, and
graphNEL, graphAM, graphBAM from graph (in Bioconductor).
The dplyr verbs in tidygraph
activate() verb from tidygraph serves as a switch between tibbles for nodes and edges. All dplyr verbs applied to tbl_graph object are applied to the active tibble.
.N() function is used to gain access to the node data while manipulating the edge data. E() will give you the edge data and .G() will give you the tbl_graph object itself.
Using tbl_graph() to build tidygraph data model.
use tbl_graph() of tinygraph package to build an tidygraph’s network graph data.frame.
GAStech_graph <- tbl_graph(nodes = GAStech_nodes,
edges = GAStech_edges_aggregated,
directed = TRUE)Reviewing the output tidygraph’s graph object
GAStech_graph# A tbl_graph: 54 nodes and 1372 edges
#
# A directed multigraph with 1 component
#
# Node Data: 54 × 4 (active)
id label Department Title
<dbl> <chr> <chr> <chr>
1 1 Mat.Bramar Administration Assistant to CEO
2 2 Anda.Ribera Administration Assistant to CFO
3 3 Rachel.Pantanal Administration Assistant to CIO
4 4 Linda.Lagos Administration Assistant to COO
5 5 Ruscella.Mies.Haber Administration Assistant to Engineering Group Mana…
6 6 Carla.Forluniau Administration Assistant to IT Group Manager
7 7 Cornelia.Lais Administration Assistant to Security Group Manager
8 44 Kanon.Herrero Security Badging Office
9 45 Varja.Lagos Security Badging Office
10 46 Stenig.Fusil Security Building Control
# ℹ 44 more rows
#
# Edge Data: 1,372 × 4
from to Weekday Weight
<int> <int> <ord> <int>
1 1 2 Sunday 5
2 1 2 Monday 2
3 1 2 Tuesday 3
# ℹ 1,369 more rowsReviewing the output tidygraph’s graph object
- The output above reveals that GAStech_graph is a tbl_graph object with 54 nodes and 4541 edges.
- The command also prints the first six rows of “Node Data” and the first three of “Edge Data”.
- It states that the Node Data is active. The notion of an active tibble within a tbl_graph object makes it possible to manipulate the data in one tibble at a time.
Changing the active object
The nodes tibble data frame is activated by default, but it can be changed with the activate() function. To rearrange the rows in the edges tibble to list those with the highest “weight” first, use activate() and then arrange().
GAStech_graph %>%
activate(edges) %>%
arrange(desc(Weight))# A tbl_graph: 54 nodes and 1372 edges
#
# A directed multigraph with 1 component
#
# Edge Data: 1,372 × 4 (active)
from to Weekday Weight
<int> <int> <ord> <int>
1 40 41 Saturday 13
2 41 43 Monday 11
3 35 31 Tuesday 10
4 40 41 Monday 10
5 40 43 Monday 10
6 36 32 Sunday 9
7 40 43 Saturday 9
8 41 40 Monday 9
9 19 15 Wednesday 8
10 35 38 Tuesday 8
# ℹ 1,362 more rows
#
# Node Data: 54 × 4
id label Department Title
<dbl> <chr> <chr> <chr>
1 1 Mat.Bramar Administration Assistant to CEO
2 2 Anda.Ribera Administration Assistant to CFO
3 3 Rachel.Pantanal Administration Assistant to CIO
# ℹ 51 more rowsPlotting Static Network Graphs with ggraph package
ggraph is an extension of ggplot2, making it easier to carry over basic ggplot skills to the design of network graphs.
There are three main aspects to a ggraph’s network graph, they are:
Plotting a basic network graph
The code chunk below uses ggraph(), geom-edge_link() and geom_node_point() to plot a network graph by using GAStech_graph.
The basic plotting function is
ggraph(), which takes the data to be used for the graph and the type of layout desired.Both of the arguments for
ggraph()are built around igraph. Therefore,ggraph()can use either an igraph object or a tbl_graph object.
ggraph(GAStech_graph) +
geom_edge_link() +
geom_node_point()
Changing the default network graph theme
Use theme_graph() to remove the x and y axes.
ggraph introduces a special ggplot theme that provides better defaults for network graphs than the normal ggplot defaults.
theme_graph(), besides removing axes, grids, and border, changes the font to Arial Narrow (this can be overridden).The ggraph theme can be set for a series of plots with the
set_graph_style()command run before the graphs are plotted or by usingtheme_graph()in the individual plots.
g <- ggraph(GAStech_graph) +
geom_edge_link(aes()) +
geom_node_point(aes())
g + theme_graph()
Changing the coloring of the plot
g <- ggraph(GAStech_graph) +
geom_edge_link(aes(colour = 'grey50')) +
geom_node_point(aes(colour = 'grey40'))
g + theme_graph(background = 'grey10',
text_colour = 'white')
Working with ggraph’s layouts
ggraph supports many standard layouts: star, circle, nicely (default), dh, gem, graphopt, grid, mds, spahere, randomly, fr, kk, drl and lgl. Figures below and on the right show layouts supported by ggraph().
Fruchterman and Reingold layout
Plot the network graph using Fruchterman and Reingold layout.
- layout argument is used to define the layout to be used.
g <- ggraph(GAStech_graph,
layout = "fr") +
geom_edge_link(aes()) +
geom_node_point(aes())
g + theme_graph()
Modifying network nodes
colour each node by referring to their respective departments
- geom_node_point is equivalent in functionality to geo_point of ggplot2. It allows for simple plotting of nodes in different shapes, colours and sizes.
g <- ggraph(GAStech_graph,
layout = "nicely") +
geom_edge_link(aes()) +
geom_node_point(aes(colour = Department,
size = 3))
g + theme_graph()
Modifying edges
The thickness of the edges will be mapped with the Weight variable.
- geom_edge_link draws edges in the simplest way - as straight lines between the start and end nodes. But, it can do more that that. In the example above, argument width is used to map the width of the line in proportional to the Weight attribute and argument alpha is used to introduce opacity on the line.
g <- ggraph(GAStech_graph,
layout = "nicely") +
geom_edge_link(aes(width=Weight),
alpha=0.2) +
scale_edge_width(range = c(0.1, 5)) +
geom_node_point(aes(colour = Department),
size = 3)
g + theme_graph()
Creating facet graphs
Another very useful feature of ggraph is faceting. In visualising network data, this technique can be used to reduce edge over-plotting in a very meaning way by spreading nodes and edges out based on their attributes. In this section, you will learn how to use faceting technique to visualise network data.
There are three functions in ggraph to implement faceting:
facet_nodes() whereby edges are only draw in a panel if both terminal nodes are present here,
facet_edges() whereby nodes are always drawn in al panels even if the node data contains an attribute named the same as the one used for the edge facetting, and
facet_graph() faceting on two variables simultaneously.
Working with facet_edges()
set_graph_style()
g <- ggraph(GAStech_graph,
layout = "nicely") +
geom_edge_link(aes(width=Weight),
alpha=0.2) +
scale_edge_width(range = c(0.1, 5)) +
geom_node_point(aes(colour = Department),
size = 2)
g + facet_edges(~Weekday)
Working with facet_edges()
uses theme() to change the position of the legend
set_graph_style()
g <- ggraph(GAStech_graph,
layout = "nicely") +
geom_edge_link(aes(width=Weight),
alpha=0.2) +
scale_edge_width(range = c(0.1, 5)) +
geom_node_point(aes(colour = Department),
size = 2) +
theme(legend.position = 'bottom')
g + facet_edges(~Weekday)
A framed facet graph
Add frame to each graph
set_graph_style()
g <- ggraph(GAStech_graph,
layout = "nicely") +
geom_edge_link(aes(width=Weight),
alpha=0.2) +
scale_edge_width(range = c(0.1, 5)) +
geom_node_point(aes(colour = Department),
size = 2)
g + facet_edges(~Weekday) +
th_foreground(foreground = "grey80",
border = TRUE) +
theme(legend.position = 'bottom')
Working with facet_nodes()
set_graph_style()
g <- ggraph(GAStech_graph,
layout = "nicely") +
geom_edge_link(aes(width=Weight),
alpha=0.2) +
scale_edge_width(range = c(0.1, 5)) +
geom_node_point(aes(colour = Department),
size = 2)
g + facet_nodes(~Department)+
th_foreground(foreground = "grey80",
border = TRUE) +
theme(legend.position = 'bottom')
Network Metrics Analysis
Computing centrality indices
Centrality measures are a collection of statistical indices use to describe the relative important of the actors are to a network. There are four well-known centrality measures, namely: degree, betweenness, closeness and eigenvector. It is beyond the scope of this hands-on exercise to cover the principles and mathematics of these measure here.
mutate() of dplyr is used to perform the computation.
the algorithm used, on the other hand, is the centrality_betweenness() of tidygraph.
g <- GAStech_graph %>%
mutate(betweenness_centrality = centrality_betweenness()) %>%
ggraph(layout = "fr") +
geom_edge_link(aes(width=Weight),
alpha=0.2) +
scale_edge_width(range = c(0.1, 5)) +
geom_node_point(aes(colour = Department,
size=betweenness_centrality))
g + theme_graph()
Visualising network metrics
From ggraph v2.0 onward tidygraph algorithms such as centrality measures can be accessed directly in ggraph calls. This means that it is no longer necessary to precompute and store derived node and edge centrality measures on the graph in order to use them in a plot.
g <- GAStech_graph %>%
ggraph(layout = "fr") +
geom_edge_link(aes(width=Weight),
alpha=0.2) +
scale_edge_width(range = c(0.1, 5)) +
geom_node_point(aes(colour = Department,
size = centrality_betweenness()))
g + theme_graph()
Visualising Community
tidygraph package inherits many of the community detection algorithms imbedded into igraph, including Edge-betweenness (group_edge_betweenness), Leading eigenvector (group_leading_eigen), Fast-greedy (group_fast_greedy), Louvain (group_louvain), Walktrap (group_walktrap), Label propagation (group_label_prop), InfoMAP (group_infomap), Spinglass (group_spinglass), and Optimal (group_optimal).
group_edge_betweenness() is used below
g <- GAStech_graph %>%
mutate(community = as.factor(group_edge_betweenness(weights = Weight, directed = TRUE))) %>%
ggraph(layout = "fr") +
geom_edge_link(aes(width=Weight),
alpha=0.2) +
scale_edge_width(range = c(0.1, 5)) +
geom_node_point(aes(colour = community))
g + theme_graph()
Building Interactive Network Graph with visNetwork
visNetwork() is a R package for network visualization, using vis.js javascript library.
visNetwork() function uses a nodes list and edges list to create an interactive graph.
The nodes list must include an “id” column, and the edge list must have “from” and “to” columns.
The function also plots the labels for the nodes, using the names of the actors from the “label” column in the node list.
Resulting graph:
The nodes can be moved and the graph will use an algorithm to keep the nodes properly spaced.
The plot can be moved around to be re-centered, zoomed in and out.
Data preparation
GAStech_edges_aggregated <- GAStech_edges %>%
left_join(GAStech_nodes, by = c("sourceLabel" = "label")) %>%
rename(from = id) %>%
left_join(GAStech_nodes, by = c("targetLabel" = "label")) %>%
rename(to = id) %>%
filter(MainSubject == "Work related") %>%
group_by(from, to) %>%
summarise(weight = n()) %>%
filter(from!=to) %>%
filter(weight > 1) %>%
ungroup()Plotting the first interactive network graph
visNetwork(GAStech_nodes,
GAStech_edges_aggregated)Working with layout
Fruchterman and Reingold layout is used.
visNetwork(GAStech_nodes,
GAStech_edges_aggregated) %>%
visIgraphLayout(layout = "layout_with_fr") Working with visual attributes - Nodes
visNetwork() looks for a field called “group” in the nodes object and colour the nodes according to the values of the group field.
rename Department field to group
GAStech_nodes <- GAStech_nodes %>%
rename(group = Department) visNetwork shades the nodes by assigning unique colour to each category in the group field.
visNetwork(GAStech_nodes,
GAStech_edges_aggregated) %>%
visIgraphLayout(layout = "layout_with_fr") %>%
visLegend() %>%
visLayout(randomSeed = 123)Working with visual attributes - Edges
visEdges() is used to symbolise the edges.
The argument arrows is used to define where to place the arrow.
The smooth argument is used to plot the edges using a smooth curve.
visNetwork(GAStech_nodes,
GAStech_edges_aggregated) %>%
visIgraphLayout(layout = "layout_with_fr") %>%
visEdges(arrows = "to",
smooth = list(enabled = TRUE,
type = "curvedCW")) %>%
visLegend() %>%
visLayout(randomSeed = 123)Interactivity
visOptions() is used to incorporate interactivity features in the data visualisation.
The argument highlightNearest highlights nearest when clicking a node.
The argument nodesIdSelection adds an id node selection creating an HTML select element.
visNetwork(GAStech_nodes,
GAStech_edges_aggregated) %>%
visIgraphLayout(layout = "layout_with_fr") %>%
visOptions(highlightNearest = TRUE,
nodesIdSelection = TRUE) %>%
visLegend() %>%
visLayout(randomSeed = 123)