Social Network Analysis with R : igraph
JungHwan Yun
[Author]
@ JungHwan Yun
@ Master Student in Data-Science
@ Seoul National University of Science & Technology(SeoulTech)
@ E-mail : junghwan.yun@seoultech.ac.kr
[Contents]
@ Topic : Social Network Analysis with R
@ Class : Social Network Analysis
@ Version : 2.0
@ Version date : 2017-05-10
@ Summary : Package “igraph”의 활용법을 다룹니다.
1. 패키지 로딩 및 작업환경 설정
- sna 패키지 설치 및 로딩
- setwd()를 이용하여 작업환경 설정
# install.packages("igraph")
#setwd("D:/Google Drive/1_MASTER_Class/3_1_Social_Network_Analysis/1_Lecture_Material/C6_R_SNA/Train_Data")
library(igraph)2. igraph package의 기본적 사용
- “igraph” 패키지를 이용한 연결망 분석
- igraph의 경우는 node와 edge를 입력데이터로 사용함.
edges <-c(1,2, 3,2, 2,4)
(g<-graph(edges, n=max(edges), directed=TRUE))## IGRAPH D--- 4 3 --
## + edges:
## [1] 1->2 3->2 2->4- vcount() : vertex(node) 개수
vcount(g)## [1] 4- ecount() : edge 개수
ecount(g)## [1] 3- is.directed() : 방향성이 있는지? T/F
is.directed(g)## [1] TRUE- get.edgelist : edge list를 행렬형태로 호출
get.edgelist(g)## [,1] [,2]
## [1,] 1 2
## [2,] 3 2
## [3,] 2 4plot(g, layout = layout.fruchterman.reingold, vertex.label=V(g)$number,edge.arrow.size=0.5)
3. Example1 : Knoke-information network
- Knoke-information 데이터를 활용한 예시
STEP1. Edge list형식으로 파일 읽어들이기
- read.graph()를 사용하여 edge파일 불러오기
(knoke.infor.1 <-read.graph("Train_Data/knoke-infor-edgelist-1.txt",format="edgelist",n=10,directed=F))## IGRAPH U--- 10 17 --
## + edges:
## [1] 1-- 2 1-- 5 2-- 3 2-- 4 2-- 5 2-- 7 2-- 8 2-- 9 3-- 5 3-- 6 3--10
## [12] 4-- 5 4-- 7 5-- 7 5-- 8 5-- 9 5--10- V()$name : vertex들의 이름을 부여
V(knoke.infor.1)$name <-c('COUN','COMM','EDUC','INDU','MAYR','WRO','NEWS','UWAY','WELF','WEST')
plot(knoke.infor.1, layout=layout.fruchterman.reingold, vertex.label=V(knoke.infor.1)$name,vertex.shape="rectangle", vertex.shape="none", vertex.label.font=2, edge.arrow.size=0.5)
# Knokemoney network
knoke.money<-read.graph("Train_Data//knoke-money-edgelist.txt",format="edgelist")
V(knoke.money)$name <-c("COUN","COMM","EDUC","INDU","MAYR","WRO","NEWS","UWAY","WELF","WEST")
plot(knoke.money, layout=layout.fruchterman.reingold, vertex.label=V(knoke.money)$name, vertex.label.font=2, edge.arrow.size=0.5)
STEP2. 네트워크 그리기
- sna패키지의 gplot()함수와 유사함. (igraph-plot() Document)[“http://igraph.org/r/doc/plot.common.html”]
- edge.arrow.size : 엣지 화살표 크기
- edge.color : 엣지 색깔
- edge.width : 엣지 두께
- vertex.color : 노드의 색상
- vertex.size : 노드의 크기
- layout : 레이아웃 형태
plot(g, edge.arrow.size= 0.5, edge.color= "gray", vertex.color= "Skyblue", vertex.size= 30, edge.width= 2, layout = layout.kamada.kawai)
STEP3. 네트워크 지표산출하기
edge.list<-c(0,3, 1,3, 2,3, 3,4, 4,5, 5,6, 6,7, 7,8) + 1
L <-graph(n=9,edge.list,directed=F)
plot(L, layout=layout.fruchterman.reingold, edge.arrow.mode="-", edge.width=2, vertex.label.cex=1.2)
- 산출지표
- degree()
- betweenness()
- closeness()
- evcent() : 위세중앙성
- centralization.degee() : degree기반 중심화
degree(L, mode="in")## [1] 1 1 1 4 2 2 2 2 1betweenness(L)## [1] 0 0 0 18 16 15 12 7 0round(closeness(L, mode="out"),2)## [1] 0.04 0.04 0.04 0.06 0.06 0.06 0.05 0.04 0.03evcent(L)$vector## [1] 0.4723926 0.4723926 0.4723926 1.0000000 0.6997053 0.4811943 0.3189268
## [8] 0.1939366 0.0916142centralization.degree(L)## $res
## [1] 1 1 1 4 2 2 2 2 1
##
## $centralization
## [1] 0.2777778
##
## $theoretical_max
## [1] 724. 네트워크 시뮬레이션
- Duncun Watz : small world network
- n = 50
- p = 0.1
- nei = 1
n <-50
p <-0.1
nei<-1
g_01 <-watts.strogatz.game(1, n, nei,p)
plot(g_01, layout=layout.circle, vertex.shape= "none", edge.arrow.size= 0.5)
(deg<-table(degree(g_01)))##
## 1 2 3
## 8 34 8- small world network
- n = 50
- p = 0.2
- nei = 1
n <-50
p <-0.2
nei<-1
# small world network
g_01 <-watts.strogatz.game(1, n, nei,p)
plot(g_01, layout=layout.circle, vertex.shape= "none", edge.arrow.size= 0.5)
(deg<-table(degree(g_01)))##
## 1 2 3 5
## 15 22 12 1- barabasi : scale-free network
- barabasi.game() function:
- power : The power of the preferential attachment,
- the default is one, ie. linear preferential attachment.
- power= 1, 1.5, 2…
barabasi.1<-barabasi.game(n=100,power=1)
plot(barabasi.1,layout=layout.fruchterman.reingold, vertex.shape="none",vertex.label.cex=.5,edge.arrow.size=.1) 
centralization.degree(barabasi.1)## $res
## [1] 35 2 6 1 3 5 2 8 2 2 2 2 4 3 1 2 1 3 2 2 2 2 2
## [24] 1 1 3 3 2 2 1 2 1 2 2 1 2 1 1 1 2 1 5 3 4 4 1
## [47] 1 1 1 1 1 1 1 2 1 2 1 1 1 2 1 1 2 1 1 1 1 1 1
## [70] 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
## [93] 1 1 1 1 1 1 1 1
##
## $centralization
## [1] 0.1684522
##
## $theoretical_max
## [1] 19602mean_distance(barabasi.1)## [1] 1.759804barabasi.2<-barabasi.game(n=100,power=2)
plot(barabasi.2,layout=layout.fruchterman.reingold, vertex.shape="none",vertex.label.cex=.5,edge.arrow.size=.1) 
centralization.degree(barabasi.2)## $res
## [1] 93 1 3 1 1 1 1 1 2 1 1 1 1 1 1 2 1 1 1 2 1 1 1
## [24] 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2
## [47] 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
## [70] 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
## [93] 1 1 1 1 1 1 1 1
##
## $centralization
## [1] 0.4643404
##
## $theoretical_max
## [1] 19602mean_distance(barabasi.2)## [1] 1.0571435. Example2
- sample_adjmatrix
- 정방행렬을 이용하여 네트워크 산출
# Generate adjacency matrix
adj_dat=read.csv("Train_Data//sample_adjmatrix.csv", header=TRUE,
row.names= 1,check.names=FALSE)
m1=as.matrix(adj_dat) # Transform data frame to matrix
m1## 23732 23778 23824 23871 58009 58098 58256
## 23732 0 1 0 1 0 1 0
## 23778 1 0 1 1 0 1 0
## 23824 0 1 0 0 0 0 0
## 23871 1 1 0 0 1 1 0
## 58009 0 0 0 1 0 1 0
## 58098 1 1 0 1 1 0 1
## 58256 0 0 0 0 0 1 0# Generate igraphdata using adjacency matrix
g1=graph.adjacency(m1,mode="undirected",weighted=NULL,diag=FALSE)
plot.igraph(g1) # Plot adjacency matrix
- sample_edgelist
- Edge list를 이용한 네트워크 산ㅊ
edge_dat=read.csv("Train_Data/sample_edgelist.csv",header=TRUE)
e1 <-as.character(edge_dat[,1]) # Extract Word Column 1
e2 <-as.character(edge_dat[,2]) # Extract Word Column 2
m2=as.matrix(data.frame(e1,e2))
# Generate igraphdata using edgelistdata matrix
g2=graph.edgelist(m2, directed = FALSE)
plot.igraph(g2)
V(g1)$name # Show vertex labels## [1] "23732" "23778" "23824" "23871" "58009" "58098" "58256"attr_dat<-read.csv("Train_Data//sample_attributes.csv", head=TRUE)
# Assign the "Sex" attribute
V(g1)$Sex <-as.character(attr_dat$Sex[match(V(g1)$name,attr_dat$Bird.ID)])
V(g1)$Sex## [1] "F" "M" "M" "F" "F" "M" "M"# assign the "Sex" attribute as the vertex color
V(g1)$color=V(g1)$Sex
V(g1)$color=gsub("F","red",V(g1)$color) # Females will be red
V(g1)$color=gsub("M","blue",V(g1)$color) # Males will be blue
plot(g1)