-
Notifications
You must be signed in to change notification settings - Fork 37
/
GraphClasses.R
598 lines (540 loc) · 22.5 KB
/
GraphClasses.R
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
#' @title Graph Classes for Network Modeling
#' @name GraphClasses
#' @rdname GraphClasses
#' @description pkgnet uses R6 classes to define and encapsulate the graph
#' models for representing package networks. These classes implement different
#' types of graphs and functionality to calculate their respective graph theory
#' measures.
#'
#' Currently the only implemented type of graph is \link{DirectedGraph}
#'
#' @section Class Constructor:
#' \describe{
#' \item{\code{new(nodes, edges)}}{
#' \itemize{
#' \item{Instantiate new object of the class.}
#' \item{\bold{Args:}}{
#' \itemize{
#' \item{\bold{\code{nodes}}: a data.table containing nodes}
#' \item{\bold{\code{edges}}: a data.table containing edges}
#' }
#' }
#' \item{\bold{Returns:}}{
#' \itemize{
#' \item{Object of the class}
#' }
#' }
#' }
#' }
#' }
#'
#' @section Public Methods:
#' \describe{
#' \item{\code{node_measures(measures = NULL)}}{
#' \itemize{
#' \item{Return specified node-level measures, calculating if necessary.
#' See Node Measures section below for details about each measure.}
#' \item{\bold{Args:}}{
#' \itemize{
#' \item{\bold{\code{measures}}: character vector of measure
#' names. Default NULL will return those that are already
#' calculated.}
#' }
#' }
#' \item{\bold{Returns:}}{
#' \itemize{
#' \item{data.table with specified node meaures as columns}
#' }
#' }
#' }
#' }
#' \item{\code{graph_measures(measures = NULL)}}{
#' \itemize{
#' \item{Return specified graph-level measures, calculating if necessary.
#' See Graph Measures section below for details about each measure.}
#' \item{\bold{Args:}}{
#' \itemize{
#' \item{\bold{\code{measures}}: character vector of measure
#' names. Default NULL will return those that are already
#' calculated.}
#' }
#' }
#' \item{\bold{Returns:}}{
#' \itemize{
#' \item{list with specified graph measures}
#' }
#' }
#' }
#' }
#' }
#'
#' @section Public Fields:
#' \describe{
#' \item{\bold{\code{nodes}}}{: node data.table, read-only}
#' \item{\bold{\code{edges}}}{: edge data.table, read-only}
#' \item{\bold{\code{igraph}}}{: igraph object, read-only}
#' \item{\bold{\code{available_node_measures}}}{: character vector of all
#' supported node measures. See Node Measures section below for detailed
#' descriptions. Read-only.}
#' \item{\bold{\code{available_graph_measures}}}{: character vector of all
#' supported graph measures. See Graph Measures section below for detailed
#' descriptions. Read-only.}
#' \item{\bold{\code{default_node_measures}}}{: character vector of default
#' node measures. See Node Measures section below for detailed descriptions.
#' Read-only.}
#' \item{\bold{\code{default_graph_measures}}}{: character vector of default
#' graph measures. See Graph Measures section below for detailed descriptions.
#' Read-only.}
#' }
#'
#'
#' @section Special Methods:
#' \describe{
#' \item{\code{clone(deep = FALSE)}}{
#' \itemize{
#' \item{Method for copying an object. See \href{https://adv-r.hadley.nz/r6.html#r6-semantics}{\emph{Advanced R}} for the intricacies of R6 reference semantics.}
#' \item{\bold{Args:}}{
#' \itemize{
#' \item{\bold{\code{deep}}: logical. Whether to recursively clone nested R6 objects.}
#' }
#' }
#' \item{\bold{Returns:}}{
#' \itemize{
#' \item{Cloned object of this class.}
#' }
#' }
#' }
#' }
#' \item{\code{print()}}{
#' \itemize{
#' \item{Print igraph object.}
#' \item{\bold{Returns:}}{
#' \itemize{
#' \item{Self}
#' }
#' }
#' }
#' }
#' }
#' @keywords internal
NULL
## Base class for Graphs
#' @importFrom R6 R6Class
#' @importFrom igraph graph.edgelist make_empty_graph vertex
#' @importFrom data.table data.table
#' @importFrom assertthat assert_that
AbstractGraph <- R6::R6Class(
classname = "AbstractGraph"
, public = list(
initialize = function(nodes, edges) {
# Input validation
assertthat::assert_that(
data.table::is.data.table(nodes)
, 'node' %in% names(nodes)
, data.table::is.data.table(edges)
, all(c('SOURCE', 'TARGET') %in% names(edges))
)
# Store pointers to node and edge data.tables
private$protected$nodes <- nodes
private$protected$edges <- edges
return(invisible(self))
}
, node_measures = function(measures = NULL){
# If not specifying, return node table
if (is.null(measures)) {
return(self$nodes)
}
assertthat::assert_that(is.character(measures))
for (m in measures) {
# Input validation
assertthat::assert_that(
all(m %in% self$available_node_measures)
, msg = sprintf('%s not in $available_node_measures()', m)
)
# If not already calculated it, calculate and add to node DT
if (!m %in% names(self$nodes)) {
log_info(sprintf("Calculating %s...", m))
result <- private$node_measure_functions[[m]](self)
resultDT <- data.table::data.table(
node_name = names(result)
, result = result
)
setkeyv(resultDT, 'node_name')
self$nodes[, eval(m) := resultDT[node, result]]
}
}
return(self$nodes[, .SD, .SDcols = c('node', measures)])
}
, graph_measures = function(measures = NULL){
# If not specifying, return full list
if (is.null(measures)) {
return(private$protected$graph_measures)
}
assertthat::assert_that(is.character(measures))
for (m in measures) {
# Input validation
assertthat::assert_that(
m %in% self$available_graph_measures
, msg = sprintf('%s not in $available_graph_measures()', m)
)
# If not already calculated, calculate
if (!m %in% names(private$protected$graph_measures)) {
log_info(sprintf("Calculating %s", m))
result <- private$graph_measure_functions[[m]](self)
private$protected$graph_measures[[m]] <- result
}
}
return(private$protected$graph_measures[measures])
}
, print = function(){
print(self$igraph)
invisible(self)
}
) # /public
, active = list(
# Read-only access to node and edge data.tables
nodes = function(){return(private$protected$nodes)}
, edges = function(){return(private$protected$edges)}
# Read-only access to igraph objects
, igraph = function(){
if (is.null(private$protected$igraph)) {
private$initialize_igraph()
}
return(private$protected$igraph)
}
, available_node_measures = function(){
return(names(private$node_measure_functions))
}
, available_graph_measures = function(){
return(names(private$graph_measure_functions))
}
, default_node_measures = function(){
log_fatal('Default node measures not implemented.')
}
, default_graph_measures = function(){
log_fatal('Default graph measures not implemented.')
}
) # /active
, private = list(
protected = list(
nodes = NULL
, edges = NULL
, igraph = NULL
, graph_measures = list()
)
, initialize_igraph = function(directed){
log_info("Constructing igraph object...")
# Connected graph
if (nrow(self$edges) > 0) {
# A graph with edges
connectedGraph <- igraph::graph.edgelist(
as.matrix(self$edges[,list(SOURCE,TARGET)])
, directed = directed
)
} else {
connectedGraph <- igraph::make_empty_graph(directed = directed)
}
# Unconnected graph
orphanNodes <- base::setdiff(
self$nodes[, node]
, unique(c(self$edges[, SOURCE], self$edges[, TARGET]))
)
unconnectedGraph <- igraph::make_empty_graph(directed = directed) + igraph::vertex(orphanNodes)
# Complete graph
completeGraph <- connectedGraph + unconnectedGraph
# Store in protected cache
private$protected$igraph <- completeGraph
log_info("...done constructing igraph object.")
return(invisible(NULL))
} # /initialize_igraph
# Functions for node measures
# All functions should return a named vector of node measure values
, node_measure_functions = list()
# Functions for graph-level measures
# All functions should return numeric of length 1
, graph_measure_functions = list()
) # /private
)
#' @title Directed Graph Network Model
#' @name DirectedGraph
#' @description R6 class defining a directed graph model for representing a
#' network, including methods to calculate various measures from graph theory.
#' The \link[igraph:igraph-package]{igraph} package is used as a backend for
#' calculations.
#' @format An \code{\link[R6]{R6Class}} generator object
#' @inheritSection GraphClasses Class Constructor
#' @inheritSection GraphClasses Public Methods
#' @inheritSection GraphClasses Public Fields
#' @inheritSection GraphClasses Special Methods
#' @inheritSection DirectedGraphMeasures Node Measures
#' @inheritSection DirectedGraphMeasures Graph Measures
#' @importFrom R6 R6Class
#' @importFrom igraph degree closeness betweenness
#' @importFrom igraph page_rank hub_score authority_score
#' @importFrom igraph neighborhood.size vcount V
#' @importFrom igraph centralize centr_degree_tmax
#' @importFrom igraph centr_clo_tmax centr_betw_tmax
DirectedGraph <- R6::R6Class(
classname = "DirectedGraph"
, inherit = AbstractGraph
, public = list(
) # / public
, active = list(
default_node_measures = function() {
return(c(
"outDegree"
, "inDegree"
, "outSubgraphSize"
, "inSubgraphSize"
, "betweenness"
, "pageRank"
))
}
, default_graph_measures = function() {
return(c(
"graphOutDegree"
, "graphInDegree"
, "graphBetweenness"
))
}
)
, private = list(
# Initialize igraph object
initialize_igraph = function() {
super$initialize_igraph(directed = TRUE)
}
# Functions for node measures
# All functions should return a named vector of node measure values
, node_measure_functions = list(
# Out-Degree
outDegree = function(self){
igraph::degree(
graph = self$igraph
, mode = "out"
, loops = TRUE
)
}
# In-Degree
, inDegree = function(self){
igraph::degree(
graph = self$igraph
, mode = "in"
, loops = TRUE
)
}
# Out-Closeness
# Closeness doesn't really work for directed graphs that are not
# strongly connected.
# igraph calculates a thing anyways and gives a warning
# Typically given as normalized values
, outCloseness = function(self){
suppressWarnings(igraph::closeness(
graph = self$igraph
, mode = "out"
, normalized = TRUE
))
}
# In-Closeness
# Closeness doesn't really work for directed graphs that are not
# strongly connected.
# igraph calculates a thing anyways and gives a warning
# Typically given as normalized values
, inCloseness = function(self){
suppressWarnings(igraph::closeness(
graph = self$igraph
, mode = "out"
, normalized = TRUE
))
}
# Out-Subgraph Size -- Rooted subgraph out from node
, outSubgraphSize = function(self){
# Calculate using out-neighborhood size with order of longest
# possible path
result <- igraph::neighborhood.size(
graph = self$igraph
, order = igraph::vcount(self$igraph)
, mode = "out"
)
# Subtract 1 so we don't include the root node itself
result <- result - 1
names(result) <- igraph::V(self$igraph)$name
return(result)
}
# In-Subgraph Size -- Rooted subgraph into node
, inSubgraphSize = function(self){
# Calculate using in-neighborhood size with order of longest
# possible path
result <- igraph::neighborhood.size(
graph = self$igraph
, order = igraph::vcount(self$igraph)
, mode = "in"
)
# Subtract 1 so we don't include the root node itself
result <- result - 1
names(result) <- igraph::V(self$igraph)$name
return(result)
}
# Betweenness
, betweenness = function(self){
igraph::betweenness(
graph = self$igraph
, directed = TRUE
)
}
# Page Rank
, pageRank = function(self){
igraph::page_rank(
graph = self$igraph
, directed = TRUE
)$vector
}
# Hub Score
, hubScore = function(self){
igraph::hub_score(
graph = self$igraph
, scale = TRUE
)$vector
}
# Authority Score
, authorityScore = function(self){
igraph::authority_score(
graph = self$igraph
, scale = TRUE
)$vector
}
) #/node_measure_functions
# Functions for graph-level measures
# All functions should return numeric of length 1
, graph_measure_functions = list(
graphOutDegree = function(self){
measure <- 'outDegree'
igraph::centralize(
scores = self$node_measures(measure)[, get(measure)]
, theoretical.max = igraph::centr_degree_tmax(
graph = self$igraph
, mode = "out"
, loops = TRUE
)
, normalized = TRUE
)
}
, graphInDegree = function(self){
measure <- 'inDegree'
igraph::centralize(
scores = self$node_measures(measure)[, get(measure)]
, theoretical.max = igraph::centr_degree_tmax(
graph = self$igraph
, mode = "in"
, loops = TRUE
)
, normalized = TRUE
)
}
, graphOutCloseness = function(self){
measure <- 'outCloseness'
igraph::centralize(
scores = self$node_measures(measure)[, get(measure)]
, theoretical.max = igraph::centr_clo_tmax(
graph = self$igraph
, mode = "out")
, normalized = TRUE
)
}
, graphInCloseness = function(self){
measure <- 'inCloseness'
igraph::centralize(
scores = self$node_measures(measure)[, get(measure)]
, theoretical.max = igraph::centr_clo_tmax(
graph = self$igraph
, mode = "in")
, normalized = TRUE
)
}
, graphBetweenness = function(self){
measure <- 'betweenness'
igraph::centralize(
scores = self$node_measures(measure)[, get(measure)]
, theoretical.max = igraph::centr_betw_tmax(
graph = self$igraph
, directed = TRUE)
, normalized = TRUE
)
}
) # /graph_measures_functions
) # /private
)
#' @title Measures for Directed Graph Class
#' @name DirectedGraphMeasures
#' @rdname DirectedGraphMeasures
#' @keywords internal
#' @section Node Measures:
#' \describe{
#' \item{\bold{\code{outDegree}}}{: outdegree, the number of outward edges (tail ends).
#' Calculated by \code{\link[igraph:degree]{igraph::degree}}.
#' [\href{https://en.wikipedia.org/wiki/Directed_graph#Indegree_and_outdegree}{Wikipedia}]}
#' \item{\bold{\code{inDegree}}}{: indegree, number of inward edges (head ends).
#' Calculated by \code{\link[igraph:degree]{igraph::degree}}.
#' [\href{https://en.wikipedia.org/wiki/Directed_graph#Indegree_and_outdegree}{Wikipedia}]}
#' \item{\bold{\code{outCloseness}}}{: closeness centrality (out), a measure of
#' path lengths to other nodes along edge directions.
#' Calculated by \code{\link[igraph:closeness]{igraph::closeness}}.
#' [\href{https://en.wikipedia.org/wiki/Closeness_centrality}{Wikipedia}]}
#' \item{\bold{\code{inCloseness}}}{: closeness centrality (in), a measure of
#' path lengths to other nodes in reverse of edge directions.
#' Calculated by \code{\link[igraph:closeness]{igraph::closeness}}.
#' [\href{https://en.wikipedia.org/wiki/Closeness_centrality}{Wikipedia}]}
#' \item{\bold{\code{outSubgraphSize}}}{: number of other nodes in the rooted
#' subgraph out from this node, i.e., count of all nodes reachable by following edges
#' out from this node.
#' Calculated by \code{\link[igraph:neighborhood.size]{igraph::neighborhood.size}}.
#' [\href{https://en.wikipedia.org/wiki/Rooted_graph}{Wikipedia}]}
#' \item{\bold{\code{inSubgraphSize}}}{: number of other nodes in the rooted
#' subgraph into this node, i.e., count all nodes reachable by following edges
#' into this node in reverse direction.
#' Calculated by \code{\link[igraph:neighborhood.size]{igraph::neighborhood.size}}.
#' [\href{https://en.wikipedia.org/wiki/Rooted_graph}{Wikipedia}]}
#' \item{\bold{\code{betweenness}}}{: betweenness centrality, a measure of
#' the number of shortest paths in graph passing through this node
#' Calculated by \code{\link[igraph:betweenness]{igraph::betweenness}}.
#' [\href{https://en.wikipedia.org/wiki/Betweenness_centrality}{Wikipedia}]}
#' \item{\bold{\code{pageRank}}}{: Google PageRank.
#' Calculated by \code{\link[igraph:page_rank]{igraph::page_rank}}.
#' [\href{https://en.wikipedia.org/wiki/PageRank}{Wikipedia}]}
#' \item{\bold{\code{hubScore}}}{: hub score from Hyperlink-Induced Topic
#' Search (HITS) algorithm.
#' Calculated by \code{\link[igraph:hub_score]{igraph::hub_score}}.
#' [\href{https://en.wikipedia.org/wiki/HITS_algorithm}{Wikipedia}]}
#' \item{\bold{\code{authorityScore}}}{: authority score from
#' Hyperlink-Induced Topic Search (HITS) algorithm.
#' Calculated by \code{\link[igraph:authority_score]{igraph::authority_score}}.
#' [\href{https://en.wikipedia.org/wiki/HITS_algorithm}{Wikipedia}]}
#' }
#' @section Graph Measures:
#' \describe{
#' \item{\bold{\code{graphOutDegree}}}{: graph freeman centralization for
#' outdegree. A measure of the most central node by outdegree in relation to
#' all other nodes.
#' Calculated by \code{\link[igraph:centralize]{igraph::centralize}}.
#' [\href{https://en.wikipedia.org/wiki/Centrality#Freeman_centralization}{Wikipedia}]}
#' \item{\bold{\code{graphInDegree}}}{: graph Freeman centralization for
#' indegree. A measure of the most central node by indegree in relation to
#' all other nodes.
#' Calculated by \code{\link[igraph:centralize]{igraph::centralize}}.
#' [\href{https://en.wikipedia.org/wiki/Centrality#Freeman_centralization}{Wikipedia}]}
#' \item{\bold{\code{graphOutClosness}}}{: graph Freeman centralization for
#' out-closeness. A measure of the most central node by out-closeness in relation to
#' all other nodes.
#' Calculated by \code{\link[igraph:centralize]{igraph::centralize}}.
#' [\href{https://en.wikipedia.org/wiki/Centrality#Freeman_centralization}{Wikipedia}]}
#' \item{\bold{\code{graphInCloseness}}}{: graph Freeman centralization for
#' outdegree. A measure of the most central node by outdegree in relation to
#' all other nodes.
#' Calculated by \code{\link[igraph:centralize]{igraph::centralize}}.
#' [\href{https://en.wikipedia.org/wiki/Centrality#Freeman_centralization}{Wikipedia}]}
#' \item{\bold{\code{graphBetweennness}}}{: graph Freeman centralization for
#' betweenness A measure of the most central node by betweenness in relation to
#' all other nodes.
#' Calculated by \code{\link[igraph:centralize]{igraph::centralize}}.
#' [\href{https://en.wikipedia.org/wiki/Centrality#Freeman_centralization}{Wikipedia}]}
#' }
NULL