Hierarchical Data: the data.tree package in R
Joseph Bolton
18 December 2019
7777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777 7777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777 7777777777777777777777777777777777777777777777777477777717777777777777777777777777777777777777777777 7777777777777777777777777777777227777777774534175527771552777777132177727777777777777777777777777777 7777777777777777777777771177775537777777777224313325274443145527165512567777777777777777777777777777 7777777777777777777777772532775422221777777777712322771432234177771113323715534777777777777777777777 7777777777777777777777771343471253557713477134154294777234355271447714722124217777777777777777777777 7777777777777777777777777124424333417715511455235210416477122775537771323455547777777777777777777777 7777777777777777777712173553134777772331245222771172802714115522411117567711122177777777777777777777 7777777777777777777775434222126715115547165574535351097265275323465317527772553125577777777777777777 7777777777777777777771455215276515572444436279111716081132771536122771033217221255377777777777777777 7777777777777777777777712214311662142135421360917772882711434139112710527237455224343531777777777777 7777777777777777777777243324339688502341154172606272882513341771057204422531724422334217777777777777 7777777777777777777723417151144114086177733324573843889435354177405021552433542325332777777777777777 7777777777777777777771221357714177740862712442242488894243441777685152124722233772435377777777777777 7777777777777777777723341442233569966888641754332788813571227716927735144723517777777777777777777777 7777777777777777777713411434172532224360888682777288522125311504777772505951777777777777777777777777 7777777777777777134442233145412132723772350809177080112252368855556600542777777777777777777777777777 7777777777777777714444142174537165214224422398043885136888654222222456621553177777777777777777777777 7777777777777777777243246426541243333342224508888880888621777777772553254712217777777777777777777777 7777777777777777773553273211126517437777777715888888861777777777777114242245353177777777777777777777 7777777777777777772217777777711714327777777777488888377777777777777723342244211777777777777777777777 7777777777777777777777777777777777777777777777788885777777777777777715317135577777777777777777777777 7777777777777777777777777777777777777777777777788882777777777777777777777772317777777777777777777777 7777777777777777777777777777777777777777777777788881777777777777777777777777777777777777777777777777 7777777777777777777777777777777777777777777777188881777777777777777777777777777777777777777777777777 7777777777777777777777777777777777777777777777188882777777777777777777777777777777777777777777777777 7777777777777777777777777777777777777777777777288883777777777777777777777777777777777777777777777777 7777777777777777777777777777777777777777777777388886777777777777777777777777777777777777777777777777 7777777777777777777777777777777777777777777777688888177777777777777777777777777777777777777777777777 7777777777777777777777777777777777777777777772888888617777777777777777777777777777777777777777777777 7777777777777777777777777777777777777777124440888888886322117777777777777777777777777777777777777777 7777777777777777777777777777777711222455668888888888880900000064177777777777777777177777777777777777 7777777777777777777777777771355606342217395468888880088932777123665553434177777774277777777777777777 7777777777777777777712112553272317777245277209886588230806669542177723471455333531777777777777777777 7777777777777777777771555277122245565947771601089758021660954112533177177774517777777777777777777777 7777777777777777777711777777777263277571222357288210804157145964727217777777277777777777777777777777 7777777777777777777777777777774677771412771037738974458427777164627777777777777777777777777777777777 7777777777777777777777777772459177772777719657728812575017777147163227777777777777777777777777777777 7777777777777777777777777771447777777777392137150813174054777777715112177777777777777777777777777777 7777777777777777777777777111777777777726477124448621776572517777771047777777777777777777777777777777 7777777777777777777777777777777777777501777747108277718577727777777252217777777777777777777777777777 7777777777777777777777777777777777776337777177980777264077777777777723177777777777777777777777777777 7777777777777777777777777777777777744227777774856177157337777777777777227777777777777777777777777777 7777777777777777777777777777777777757717777710822577247745177777777777771777777777777777777777777777 7777777777777777777777777777777777227777777155817577227772617777777777777777777777777777777777777777 7777777777777777777777777777777777117777777542817477717777267777777777777777777777777777777777777777 7777777777777777777777777777777777777777772417047177777777762777777777777777777777777777777777777777 7777777777777777777777777777777777777777771277897777777777732777777777777777777777777777777777777777 7777777777777777777777777777777777777777772777654777777777751777777777777777777777777777777777777777 7777777777777777777777777777777777777777777777375177777777117777777777777777777777777777777777777777 7777777777777777777777777777777777777777777777477377777777777777777777777777777777777777777777777777 7777777777777777777777777777777777777777777777777147777777777777777777777777777777777777777777777777 7777777777777777777777777777777777777777777777777747777777777777777777777777777777777777777777777777 7777777777777777777777777777777777777777777777777717777777777777777777777777777777777777777777777777 7777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777
This post is a brief exploration of the functionality of the data.tree package, which is used for working with data that has a hierarchical structure.
Here is my setup, with the packages that I’ve used:
library(tidyverse)
library(data.tree)
library(treemap) # for the dataset GNI2014
library(DiagrammeR) # for nice plots of the trees
library(RColorBrewer)
library(igraph)
library(networkD3)
library(RColorBrewer)
# change formatting of code output:
knitr::opts_chunk$set(
class.output = "bg-primary",
class.message = "bg-info text-info",
class.warning = "bg-warning text-warning",
class.error = "bg-danger text-danger"
)Resources
To make this post, I’ve extracted the examples and bits of information that were most useful to me from the following two sources:
https://cran.r-project.org/web/packages/data.tree/vignettes/data.tree.html
https://cran.r-project.org/web/packages/data.tree/vignettes/applications.html
These resources are clear and thorough, and I couldn’t recommend them more highly.
Creating a data.tree from scratch
First, we create a data.tree structure from scratch.
This example comes from https://cran.r-project.org/web/packages/data.tree/vignettes/data.tree.html
Create the root node:
## levelName
## 1 Acme Inc.Create 3 children of the root node: (accounting, research and IT)
accounting <- acme$AddChild("Accounting")
research <- acme$AddChild("Research")
it <- acme$AddChild("IT")
acme %>%
as.data.frame() %>%
as.matrix() %>%
print(quote=FALSE)## levelName
## [1,] Acme Inc.
## [2,] ¦--Accounting
## [3,] ¦--Research
## [4,] °--ITGive the accounting node 2 children:
software <- accounting$AddChild("New Software")
standards <- accounting$AddChild("New Accounting Standards")
acme %>%
as.data.frame() %>%
as.matrix() %>%
print(quote=FALSE) ## levelName
## [1,] Acme Inc.
## [2,] ¦--Accounting
## [3,] ¦ ¦--New Software
## [4,] ¦ °--New Accounting Standards
## [5,] ¦--Research
## [6,] °--ITGive the Research node 2 children:
newProductLine <- research$AddChild("New Product Line")
newLabs <- research$AddChild("New Labs")
acme %>%
as.data.frame() %>%
as.matrix() %>%
print(quote=FALSE) ## levelName
## [1,] Acme Inc.
## [2,] ¦--Accounting
## [3,] ¦ ¦--New Software
## [4,] ¦ °--New Accounting Standards
## [5,] ¦--Research
## [6,] ¦ ¦--New Product Line
## [7,] ¦ °--New Labs
## [8,] °--ITGive the IT node 3 children:
outsource <- it$AddChild("Outsource")
agile <- it$AddChild("Go agile")
goToR <- it$AddChild("Switch to R")
acme %>%
as.data.frame() %>%
as.matrix() %>%
print(quote=FALSE) ## levelName
## [1,] Acme Inc.
## [2,] ¦--Accounting
## [3,] ¦ ¦--New Software
## [4,] ¦ °--New Accounting Standards
## [5,] ¦--Research
## [6,] ¦ ¦--New Product Line
## [7,] ¦ °--New Labs
## [8,] °--IT
## [9,] ¦--Outsource
## [10,] ¦--Go agile
## [11,] °--Switch to RClimbing the tree
We can consider only the IT branch of the tree:
## levelName
## [1,] IT
## [2,] ¦--Outsource
## [3,] ¦--Go agile
## [4,] °--Switch to RWe can consider only the ‘Switch to R’ branch of the IT branch of the tree:
## levelName
## [1,] Switch to RThe children nodes of a node can be accessed using the syntax .$children[[i]]:
## levelName
## [1,] Accounting
## [2,] ¦--New Software
## [3,] °--New Accounting Standards## levelName
## [1,] Research
## [2,] ¦--New Product Line
## [3,] °--New Labs## levelName
## 1 New Accounting StandardsNow, we add values (cost and probability) to the leaves of the tree (the leaves are the terminal nodes):
acme$Accounting$`New Software`$cost <- 1000000
acme$Accounting$`New Accounting Standards`$cost <- 500000
acme$Research$`New Product Line`$cost <- 2000000
acme$Research$`New Labs`$cost <- 750000
acme$IT$Outsource$cost <- 400000
acme$IT$`Go agile`$cost <- 250000
acme$IT$`Switch to R`$cost <- 50000
acme$Accounting$`New Software`$p <- 0.5
acme$Accounting$`New Accounting Standards`$p <- 0.75
acme$Research$`New Product Line`$p <- 0.25
acme$Research$`New Labs`$p <- 0.9
acme$IT$Outsource$p <- 0.2
acme$IT$`Go agile`$p <- 0.05
acme$IT$`Switch to R`$p <- 1
print(acme, "cost", "p")## levelName cost p
## 1 Acme Inc. NA NA
## 2 ¦--Accounting NA NA
## 3 ¦ ¦--New Software 1000000 0.50
## 4 ¦ °--New Accounting Standards 500000 0.75
## 5 ¦--Research NA NA
## 6 ¦ ¦--New Product Line 2000000 0.25
## 7 ¦ °--New Labs 750000 0.90
## 8 °--IT NA NA
## 9 ¦--Outsource 400000 0.20
## 10 ¦--Go agile 250000 0.05
## 11 °--Switch to R 50000 1.00We can use a recursive function, applied to each node, to sum the cost for each node across all of it’s children:
# define the cost-summing function:
cost_ftn <-
function(node)
{
result <- node$cost
if(length(result) == 0) result <- sum( sapply( node$children, cost_ftn) )
return (result)
}
# apply the function to all of the nodes:
acme$Do(
function(node)
{
node$sum_cost <- cost_ftn(node)
},
filterFun = isNotLeaf # don't apply the function to the leaves
# this is a built-in function, but we can supply any function here, including one we've created
)
print(acme, "cost", "sum_cost")## levelName cost sum_cost
## 1 Acme Inc. NA 4950000
## 2 ¦--Accounting NA 1500000
## 3 ¦ ¦--New Software 1000000 NA
## 4 ¦ °--New Accounting Standards 500000 NA
## 5 ¦--Research NA 2750000
## 6 ¦ ¦--New Product Line 2000000 NA
## 7 ¦ °--New Labs 750000 NA
## 8 °--IT NA 700000
## 9 ¦--Outsource 400000 NA
## 10 ¦--Go agile 250000 NA
## 11 °--Switch to R 50000 NAacme$Do(
function(node)
{
node$cost_all <- cost_ftn(node)
}
)
print(acme, "cost", "sum_cost", "cost_all")## levelName cost sum_cost cost_all
## 1 Acme Inc. NA 4950000 4950000
## 2 ¦--Accounting NA 1500000 1500000
## 3 ¦ ¦--New Software 1000000 NA 1000000
## 4 ¦ °--New Accounting Standards 500000 NA 500000
## 5 ¦--Research NA 2750000 2750000
## 6 ¦ ¦--New Product Line 2000000 NA 2000000
## 7 ¦ °--New Labs 750000 NA 750000
## 8 °--IT NA 700000 700000
## 9 ¦--Outsource 400000 NA 400000
## 10 ¦--Go agile 250000 NA 250000
## 11 °--Switch to R 50000 NA 50000Plotting
See https://graphviz.gitlab.io/_pages/doc/info/attrs.html for more information on the styling of data.tree plots.
Plotting of data.tree objects using the plot() function in R calls the render_graph() function from the DiagrammeR package.
Run ?DiagrammeR::render_graph in console for more information.
Here are some example plots of our tree:
SetGraphStyle( acme,
bgcolor = "black" # make the treeplot have a black background
)
SetNodeStyle( acme,
fontcolor = "white", # make text in nodes to white
color = "white" # make node outlines white
)
SetEdgeStyle( acme,
color = "white" # make edges (arrows) white
)
plot( acme,
width = 900 # specify width in pixels
)We can closely control the text appearing inside the nodes using a custom function. This could just as easily be done for text on edges too. We put on each node the total cost across all of it’s children:
GetNodeLabel <- function(node)
{
paste0( node$name,
"\n",
"$ ",
format(node$cost_all, scientific = FALSE, big.mark = ",")
)
}
print( acme, "cost", "sum_cost", "cost_all" )## levelName cost sum_cost cost_all
## 1 Acme Inc. NA 4950000 4950000
## 2 ¦--Accounting NA 1500000 1500000
## 3 ¦ ¦--New Software 1000000 NA 1000000
## 4 ¦ °--New Accounting Standards 500000 NA 500000
## 5 ¦--Research NA 2750000 2750000
## 6 ¦ ¦--New Product Line 2000000 NA 2000000
## 7 ¦ °--New Labs 750000 NA 750000
## 8 °--IT NA 700000 700000
## 9 ¦--Outsource 400000 NA 400000
## 10 ¦--Go agile 250000 NA 250000
## 11 °--Switch to R 50000 NA 50000SetNodeStyle( acme,
label = GetNodeLabel,
fontname = "helvetica",
fontcolor = "white", # make text in nodes to white
color = "white" # make node outlines white
)
plot( acme,
width = 900
)We can change the direction/orientation with the rankdir argument:
SetGraphStyle( acme,
bgcolor = "black", # make the treeplot have a black background
rankdir = "LR"
)
plot( acme )Here is an example where we create a function to dynamically choose the border colour of each node according to the cost:
# this command displays all of the palette options in RColorBrewer:
# display.brewer.all()
# get a list of all of the cost values in the whole tree:
all_cost_values <- acme$Get('cost_all', traversal = "post-order")
all_cost_values## New Software New Accounting Standards Accounting
## 1000000 500000 1500000
## New Product Line New Labs Research
## 2000000 750000 2750000
## Outsource Go agile Switch to R
## 400000 250000 50000
## IT Acme Inc.
## 700000 4950000# define the chosen palette:
define_colours <- brewer.pal(9, "Blues")
define_palette <- colorRampPalette( define_colours )
# rank the cost values:
col_order <- findInterval( all_cost_values, sort(all_cost_values) )
# define the function which returns the colour, given a cost value:
get_node_colour_ftn <- function(node)
{
define_palette( length(all_cost_values) )[ col_order[ which(all_cost_values==node$cost_all) ] ]
}
SetNodeStyle( acme,
label = GetNodeLabel,
color = get_node_colour_ftn,
penwidth = 5,
fontname = "helvetica",
fontcolor = "white", # make text in nodes to white
color = "white" # make node outlines white
)
plot( acme )Here are some alternative ways to plot the data using other packages:
Converting a data.tree to an R data.frame or list
In order to do predictive modelling or analysis, it is useful to be able to convert the information in the data.tree structure to an R data.frame or list.
What follows below are a few different ways to do this:
Another way is:
Another way is:
..or the data.tree can be converted to a nested list:
## List of 6
## $ name : chr "Acme Inc."
## $ cost_all : num 4950000
## $ sum_cost : num 4950000
## $ Accounting:List of 4
## ..$ cost_all : num 1500000
## ..$ sum_cost : num 1500000
## ..$ New Software :List of 3
## .. ..$ cost : num 1e+06
## .. ..$ cost_all: num 1e+06
## .. ..$ p : num 0.5
## ..$ New Accounting Standards:List of 3
## .. ..$ cost : num 5e+05
## .. ..$ cost_all: num 5e+05
## .. ..$ p : num 0.75
## $ Research :List of 4
## ..$ cost_all : num 2750000
## ..$ sum_cost : num 2750000
## ..$ New Product Line:List of 3
## .. ..$ cost : num 2e+06
## .. ..$ cost_all: num 2e+06
## .. ..$ p : num 0.25
## ..$ New Labs :List of 3
## .. ..$ cost : num 750000
## .. ..$ cost_all: num 750000
## .. ..$ p : num 0.9
## $ IT :List of 5
## ..$ cost_all : num 7e+05
## ..$ sum_cost : num 7e+05
## ..$ Outsource :List of 3
## .. ..$ cost : num 4e+05
## .. ..$ cost_all: num 4e+05
## .. ..$ p : num 0.2
## ..$ Go agile :List of 3
## .. ..$ cost : num 250000
## .. ..$ cost_all: num 250000
## .. ..$ p : num 0.05
## ..$ Switch to R:List of 3
## .. ..$ cost : num 50000
## .. ..$ cost_all: num 50000
## .. ..$ p : num 1Labelling the levels of the tree
## levelName
## 1 Acme Inc.
## 2 ¦--Accounting
## 3 ¦ ¦--New Software
## 4 ¦ °--New Accounting Standards
## 5 ¦--Research
## 6 ¦ ¦--New Product Line
## 7 ¦ °--New Labs
## 8 °--IT
## 9 ¦--Outsource
## 10 ¦--Go agile
## 11 °--Switch to Radd 2 children to the Outsource node (Outsource is a child of IT):
## levelName
## 1 Acme Inc.
## 2 ¦--Accounting
## 3 ¦ ¦--New Software
## 4 ¦ °--New Accounting Standards
## 5 ¦--Research
## 6 ¦ ¦--New Product Line
## 7 ¦ °--New Labs
## 8 °--IT
## 9 ¦--Outsource
## 10 ¦ ¦--India
## 11 ¦ °--Poland
## 12 ¦--Go agile
## 13 °--Switch to Riterate through the nodes in the acme tree,
## Acme Inc. Accounting New Software
## 1 2 3
## New Accounting Standards Research New Product Line
## 3 2 3
## New Labs IT Outsource
## 3 2 3
## India Poland Go agile
## 4 4 3
## Switch to R
## 3acme$Set( type = case_when( extract_levels==1 ~ "company (root)",
extract_levels==2 ~ "department",
extract_levels==3 ~ "project",
extract_levels==4 ~ "outsource_country",
TRUE ~ "error"
)
)
print( acme, "level", "type", "cost" )## levelName level type cost
## 1 Acme Inc. 1 company (root) NA
## 2 ¦--Accounting 2 department NA
## 3 ¦ ¦--New Software 3 project 1000000
## 4 ¦ °--New Accounting Standards 3 project 500000
## 5 ¦--Research 2 department NA
## 6 ¦ ¦--New Product Line 3 project 2000000
## 7 ¦ °--New Labs 3 project 750000
## 8 °--IT 2 department NA
## 9 ¦--Outsource 3 project 400000
## 10 ¦ ¦--India 4 outsource_country NA
## 11 ¦ °--Poland 4 outsource_country NA
## 12 ¦--Go agile 3 project 250000
## 13 °--Switch to R 3 project 50000There are many ways to filter, prune and aggregate data.trees (see https://cran.r-project.org/web/packages/data.tree/vignettes/data.tree.html).
creating a data.tree from a data.frame
# this code taken from https://cran.r-project.org/web/packages/data.tree/vignettes/data.tree.html
data(GNI2014)
head(GNI2014)We specify the tree structure by creating a column called pathString:
GNI2014$pathString <- paste( "world",
GNI2014$continent,
GNI2014$country,
sep = "/"
)
GNI2014 %>% select( pathString, continent, country ) %>% head(10) %>% knitr::kable()| pathString | continent | country | |
|---|---|---|---|
| 3 | world/North America/Bermuda | North America | Bermuda |
| 4 | world/Europe/Norway | Europe | Norway |
| 5 | world/Asia/Qatar | Asia | Qatar |
| 6 | world/Europe/Switzerland | Europe | Switzerland |
| 7 | world/Asia/Macao SAR, China | Asia | Macao SAR, China |
| 8 | world/Europe/Luxembourg | Europe | Luxembourg |
| 10 | world/Oceania/Australia | Oceania | Australia |
| 11 | world/Europe/Sweden | Europe | Sweden |
| 12 | world/Europe/Denmark | Europe | Denmark |
| 14 | world/North America/United States | North America | United States |
## levelName iso3 population GNI
## 1 world NA NA
## 2 ¦--North America NA NA
## 3 ¦ ¦--Bermuda BMU 67837 106140
## 4 ¦ ¦--United States USA 313973000 55200
## 5 ¦ ¦--Canada CAN 33487208 51630
## 6 ¦ ¦--Bahamas, The BHS 309156 20980
## 7 ¦ ¦--Trinidad and Tobago TTO 1310000 20070
## 8 ¦ ¦--Puerto Rico PRI 3971020 19310
## 9 ¦ ¦--Barbados BRB 284589 15310
## 10 ¦ ¦--St. Kitts and Nevis KNA 40131 14920
## 11 ¦ ¦--Antigua and Barbuda ATG 85632 13300
## 12 ¦ ¦--Panama PAN 3360474 11130
## 13 ¦ ¦--Costa Rica CRI 4253877 10120
## 14 ¦ ¦--Mexico MEX 111211789 9870
## 15 ¦ ¦--Grenada GRD 90739 7910
## 16 ¦ ¦--St. Lucia LCA 160267 7260
## 17 ¦ ¦--Dominica DMA 72660 6930
## 18 ¦ ¦--St. Vincent and the Grenadines VCT 104574 6610
## 19 ¦ ¦--Dominican Republic DOM 9650054 6040
## 20 ¦ °--... 7 nodes w/ 0 sub NA NA
## 21 °--... 6 nodes w/ 171 sub NA NAShowcasing some tree-viewing options:
## levelName
## 1 world
## 2 ¦--North America
## 3 ¦ ¦--Bermuda
## 4 ¦ ¦--United States
## 5 ¦ ¦--Canada
## 6 ¦ ¦--Bahamas, The
## 7 ¦ ¦--Trinidad and Tobago
## 8 ¦ ¦--Puerto Rico
## 9 ¦ ¦--Barbados
## 10 ¦ ¦--St. Kitts and Nevis
## 11 ¦ ¦--Antigua and Barbuda
## 12 ¦ ¦--Panama
## 13 ¦ ¦--Costa Rica
## 14 ¦ ¦--Mexico
## 15 ¦ ¦--Grenada
## 16 ¦ ¦--St. Lucia
## 17 ¦ ¦--Dominica
## 18 ¦ ¦--St. Vincent and the Grenadines
## 19 ¦ ¦--Dominican Republic
## 20 ¦ ¦--Jamaica
## 21 ¦ ¦--Belize
## 22 ¦ ¦--El Salvador
## 23 ¦ ¦--Guatemala
## 24 ¦ ¦--Honduras
## 25 ¦ ¦--Nicaragua
## 26 ¦ °--Haiti
## 27 ¦--Europe
## 28 ¦ ¦--Norway
## 29 ¦ ¦--Switzerland
## 30 ¦ ¦--Luxembourg
## 31 ¦ ¦--Sweden
## 32 ¦ ¦--Denmark
## 33 ¦ ¦--Netherlands
## 34 ¦ ¦--Austria
## 35 ¦ ¦--Finland
## 36 ¦ ¦--Germany
## 37 ¦ ¦--Iceland
## 38 ¦ ¦--Belgium
## 39 ¦ ¦--Ireland
## 40 ¦ ¦--United Kingdom
## 41 ¦ ¦--France
## 42 ¦ ¦--Andorra
## 43 ¦ ¦--Italy
## 44 ¦ ¦--Spain
## 45 ¦ ¦--Slovenia
## 46 ¦ ¦--Greece
## 47 ¦ ¦--Portugal
## 48 ¦ ¦--Malta
## 49 ¦ ¦--Estonia
## 50 ¦ ¦--Czech Republic
## 51 ¦ ¦--Slovak Republic
## 52 ¦ ¦--Lithuania
## 53 ¦ ¦--Latvia
## 54 ¦ ¦--Poland
## 55 ¦ ¦--Hungary
## 56 ¦ ¦--Russian Federation
## 57 ¦ ¦--Croatia
## 58 ¦ ¦--Romania
## 59 ¦ ¦--Bulgaria
## 60 ¦ ¦--Belarus
## 61 ¦ ¦--Montenegro
## 62 ¦ ¦--Serbia
## 63 ¦ ¦--Macedonia, FYR
## 64 ¦ ¦--Bosnia and Herzegovina
## 65 ¦ ¦--Albania
## 66 ¦ ¦--Kosovo
## 67 ¦ ¦--Ukraine
## 68 ¦ °--Moldova
## 69 ¦--Asia
## 70 ¦ ¦--Qatar
## 71 ¦ ¦--Macao SAR, China
## 72 ¦ ¦--Singapore
## 73 ¦ ¦--Kuwait
## 74 ¦ ¦--United Arab Emirates
## 75 ¦ ¦--Japan
## 76 ¦ ¦--Hong Kong SAR, China
## 77 ¦ ¦--Brunei Darussalam
## 78 ¦ ¦--Israel
## 79 ¦ ¦--Korea, Rep.
## 80 ¦ ¦--Cyprus
## 81 ¦ ¦--Saudi Arabia
## 82 ¦ ¦--Bahrain
## 83 ¦ ¦--Oman
## 84 ¦ ¦--Kazakhstan
## 85 ¦ ¦--Malaysia
## 86 ¦ ¦--Turkey
## 87 ¦ ¦--Lebanon
## 88 ¦ ¦--Turkmenistan
## 89 ¦ ¦--Azerbaijan
## 90 ¦ ¦--China
## 91 ¦ ¦--Iraq
## 92 ¦ ¦--Iran, Islamic Rep.
## 93 ¦ ¦--Thailand
## 94 ¦ ¦--Jordan
## 95 ¦ ¦--Mongolia
## 96 ¦ ¦--Armenia
## 97 ¦ ¦--Georgia
## 98 ¦ ¦--Indonesia
## 99 ¦ ¦--Philippines
## 100 ¦ °--... 17 nodes w/ 0 sub
## 101 °--... 4 nodes w/ 93 sub## levelName
## 1 world
## 2 ¦--North America
## 3 ¦ ¦--Bermuda
## 4 ¦ ¦--United States
## 5 ¦ ¦--Canada
## 6 ¦ ¦--Bahamas, The
## 7 ¦ ¦--Trinidad and Tobago
## 8 ¦ °--... 19 nodes w/ 0 sub
## 9 °--... 6 nodes w/ 183 sub## levelName
## 1 world
## 2 ¦--North America
## 3 ¦ ¦--Bermuda
## 4 ¦ °--... 23 nodes w/ 0 sub
## 5 °--... 6 nodes w/ 187 sub