데이터 과학자와 함께 하는 제19대 대통령 선거
페북 프로파일(Profile) 시각화
페북 친구 네트워크 시각화 1 2
페북 친구 네트워크를 시각화하는 코드는 지속적으로 진화를 해왔다. igraph를 활용하여 정적 네트워크 그래프 시각화 방법을 지나서,
rthreejs를 활용하여 웹브라우져에서도 정말 가볍게 네트워크 시각화가 가능하게 되었다.
getNetwork()함수를 활용해서 본인 네트워크 정보를 가져온다.graph_from_adjacency_matrix함수로 인접행렬(Adjacent Matrix)을igraph객체로 변환한다.igraph함수degree정보를 추가한다.
- 3D 자바스크립트 three.js를 R에서 구현한
threejs팩키지graphjs함수를 활용하여 3차원 시각화한다.
# 0. 환경설정 ---------------------------------------------------
## 0.1. 팩키지 불러오기
## 0.2. 토큰 설정
me_token <- "EAACEdEose0cBAFJbE5lwlt8s9lbZAmx5Nsrw5W4FgXraBViIqHi4L4MYXJ6bVlhYo4HG2ZBwyK8ZCTzr5bqLLK9xeZA1qeRHj8XmwOU1rDNM2zOAEmiZAQZBHgD3HYZChf9xEWZBoaZBpEpT2O8zEZChsjoMLkByZCXQqAg9XHgxFGp2FlcClZBl6MwGJg1gIkvK6lQZD"
# 1. 데이터 긁어오기 ---------------------------------------------------
my_network <- getNetwork(token = me_token, format="adj.matrix")Only friends who use the application will be returned
See ?getFriends for more details
|
| | 0%
|
|= | 2%
|
|=== | 4%
|
|==== | 6%
|
|===== | 8%
|
|====== | 10%
|
|======== | 12%
|
|========= | 14%
|
|========== | 16%
|
|=========== | 18%
|
|============= | 20%
|
|============== | 22%
|
|=============== | 24%
|
|================= | 25%
|
|================== | 27%
|
|=================== | 29%
|
|==================== | 31%
|
|====================== | 33%
|
|======================= | 35%
|
|======================== | 37%
|
|========================= | 39%
|
|=========================== | 41%
|
|============================ | 43%
|
|============================= | 45%
|
|=============================== | 47%
|
|================================ | 49%
|
|================================= | 51%
|
|================================== | 53%
|
|==================================== | 55%
|
|===================================== | 57%
|
|====================================== | 59%
|
|======================================== | 61%
|
|========================================= | 63%
|
|========================================== | 65%
|
|=========================================== | 67%
|
|============================================= | 69%
|
|============================================== | 71%
|
|=============================================== | 73%
|
|================================================ | 75%
|
|================================================== | 76%
|
|=================================================== | 78%
|
|==================================================== | 80%
|
|====================================================== | 82%
|
|======================================================= | 84%
|
|======================================================== | 86%
|
|========================================================= | 88%
|
|=========================================================== | 90%
|
|============================================================ | 92%
|
|============================================================= | 94%
|
|============================================================== | 96%
|
|================================================================ | 98%
|
|=================================================================| 100%
# 2. 시각화 위한 데이터 변환 ---------------------------------------------------
# 인접행렬(Adjacent Matrix)를 igraph 객체로 변환
my_network_g <- graph_from_adjacency_matrix(my_network, weighted=TRUE)
my_network_g$degree <- degree(my_network_g)
# 3. 시각화 ---------------------------------------------------
# threejs 동적 페이스북 그래프
graphjs(my_network_g,
layout=layout_with_fr(my_network_g, dim=3),
vertex.label.cex=0.5,
edge.arrow.size=1,
edge.curved=TRUE,
vertex.size = jsonlite::toJSON(my_network_g$degree/20),
vertex.color="darkkhaki",
edge.color="darkgrey",
vertex.label=V(my_network_g)$name)# 애니메이션
# graphjs(my_network_g,
# layout=list(layout_randomly(my_network_g, dim=3),
# layout_on_sphere(my_network_g),
# layout_with_drl(my_network_g, dim=3),
# layout_with_fr(my_network_g, dim=3)),
# main=list("random layout", "sphere layout", "drl layout", "fr layout"), fpl=300,
# vertex.label.cex=0.5,
# edge.arrow.size=3,
# edge.curved=TRUE,
# vertex.size = jsonlite::toJSON(my_network_g$degree/20),
# vertex.color="darkkhaki",
# edge.color="darkgrey",
# vertex.label=V(my_network_g)$name)페북 친구 프로파일 사진 시각화 3
페북 친구 네트워크를 친구 프로파일에 있는 사진을 노드에 넣어 시각화하는 것도 상기와 유사한 과정을 거치게 된다. 다만, 친구 사진 다운로드하는 과정이 추가되고 이를 Gephi, igraph, qgraph 등에 담는 과정이 추가될 따름이다.
# 0. 환경설정 ---------------------------------------------------
## 0.1. 팩키지 불러오기
#library(png)
#library(jpeg)
#library(igraph)
## 0.2. 토큰 설정
me_token <- "EAACEdEose0cBAFJbE5lwlt8s9lbZAmx5Nsrw5W4FgXraBViIqHi4L4MYXJ6bVlhYo4HG2ZBwyK8ZCTzr5bqLLK9xeZA1qeRHj8XmwOU1rDNM2zOAEmiZAQZBHgD3HYZChf9xEWZBoaZBpEpT2O8zEZChsjoMLkByZCXQqAg9XHgxFGp2FlcClZBl6MwGJg1gIkvK6lQZD"
# 1. 데이터 긁어오기 ---------------------------------------------------
## 1.1. 친구정보 가져오기
my_friends <- getFriends(token=me_token, simplify = FALSE)Only friends who use the application will be returned
See ?getFriends for more details
## 1.2. 친구네트워크 가져오기
my_network <- getNetwork(token = me_token, format="adj.matrix") + 0Only friends who use the application will be returned
See ?getFriends for more details
|
| | 0%
|
|= | 2%
|
|=== | 4%
|
|==== | 6%
|
|===== | 8%
|
|====== | 10%
|
|======== | 12%
|
|========= | 14%
|
|========== | 16%
|
|=========== | 18%
|
|============= | 20%
|
|============== | 22%
|
|=============== | 24%
|
|================= | 25%
|
|================== | 27%
|
|=================== | 29%
|
|==================== | 31%
|
|====================== | 33%
|
|======================= | 35%
|
|======================== | 37%
|
|========================= | 39%
|
|=========================== | 41%
|
|============================ | 43%
|
|============================= | 45%
|
|=============================== | 47%
|
|================================ | 49%
|
|================================= | 51%
|
|================================== | 53%
|
|==================================== | 55%
|
|===================================== | 57%
|
|====================================== | 59%
|
|======================================== | 61%
|
|========================================= | 63%
|
|========================================== | 65%
|
|=========================================== | 67%
|
|============================================= | 69%
|
|============================================== | 71%
|
|=============================================== | 73%
|
|================================================ | 75%
|
|================================================== | 76%
|
|=================================================== | 78%
|
|==================================================== | 80%
|
|====================================================== | 82%
|
|======================================================= | 84%
|
|======================================================== | 86%
|
|========================================================= | 88%
|
|=========================================================== | 90%
|
|============================================================ | 92%
|
|============================================================= | 94%
|
|============================================================== | 96%
|
|================================================================ | 98%
|
|=================================================================| 100%
my_network_el <- as.data.frame(getNetwork(token = me_token, format = "edgelist"))Only friends who use the application will be returned
See ?getFriends for more details
|
| | 0%
|
|= | 2%
|
|=== | 4%
|
|==== | 6%
|
|===== | 8%
|
|====== | 10%
|
|======== | 12%
|
|========= | 14%
|
|========== | 16%
|
|=========== | 18%
|
|============= | 20%
|
|============== | 22%
|
|=============== | 24%
|
|================= | 25%
|
|================== | 27%
|
|=================== | 29%
|
|==================== | 31%
|
|====================== | 33%
|
|======================= | 35%
|
|======================== | 37%
|
|========================= | 39%
|
|=========================== | 41%
|
|============================ | 43%
|
|============================= | 45%
|
|=============================== | 47%
|
|================================ | 49%
|
|================================= | 51%
|
|================================== | 53%
|
|==================================== | 55%
|
|===================================== | 57%
|
|====================================== | 59%
|
|======================================== | 61%
|
|========================================= | 63%
|
|========================================== | 65%
|
|=========================================== | 67%
|
|============================================= | 69%
|
|============================================== | 71%
|
|=============================================== | 73%
|
|================================================ | 75%
|
|================================================== | 76%
|
|=================================================== | 78%
|
|==================================================== | 80%
|
|====================================================== | 82%
|
|======================================================= | 84%
|
|======================================================== | 86%
|
|========================================================= | 88%
|
|=========================================================== | 90%
|
|============================================================ | 92%
|
|============================================================= | 94%
|
|============================================================== | 96%
|
|================================================================ | 98%
|
|=================================================================| 100%
## 1.3. 친구 사진 다운로드 -----------------------------------------------------------------
dir.create("FbImages")
my_friends$picture.dld <- NA
for (i in 1:dim(my_friends)[1]) {
# Fb images 프로파일 사진 거의 대부분이 JPEG
pic_ext <- ".jpg"
if(grepl(".gif$", my_friends$picture[i])) pic_ext <- ".gif"
# 페북사용자 이름에 파일확장자 추가
my_friends$picture.dld[i] <- paste0("FBImages/", sub(" ", "_", my_friends$name[i]), pic_ext)
# UTF-8 인코딩을 갖지 않는 사용자 처리로직:
if (Encoding(my_friends$name[i])=="UTF-8") {
my_friends$picture.dld[i] <- paste0("FBImages/", "FbID_", my_friends$id[i], pic_ext) }
download.file(my_friends$picture[i], my_friends$picture.dld[i], mode = 'wb')
}
# 사진 정보 저장
colnames(my_network_el) <- c("Source", "Target")
my_friends$ID <- my_friends$name
my_friends$image <- gsub("FBImages/(*.)", "\\1", my_friends$picture.dld)
# write.csv(my_network_el, file="Facebook-friend-EDGES.csv", row.names=F)
# write.csv(my_friends, file="Facebook-friend-NODES.csv", row.names=F)
# 2. 네트워크 데이터 준비 ---------------------------------------------------
fb_net <- graph.adjacency(my_network)
my_friends$degree <- degree(fb_net)
my_friends$color <- "gray45"
my_friends$color[my_friends$gender=="female"] <- "lightpink3"
my_friends$color[my_friends$gender=="male"] <- "lightblue"
my_network_el$color <- my_friends$color[match(my_network_el$Source, my_friends$name)]시각화를 위해 사진이 들어가지 않는 그래프와 사진이 들어간 그래프를 넣어 시각화한다.
# 3. 페북 프로파일 사진 시각화 ---------------------------------------------------
## 3.1. 사진없이 시각화
l <- layout.fruchterman.reingold(fb_net, niter=10000,
area=vcount(fb_net)^2.3,
repulserad=vcount(fb_net)^2.65)Warning in layout_with_fr(structure(list(52, TRUE, c(0, 0, 0, 0, 0, 0, 0, :
Argument `area' is deprecated and has no effect
Warning in layout_with_fr(structure(list(52, TRUE, c(0, 0, 0, 0, 0, 0, 0, :
Argument `repulserad' is deprecated and has no effect
plot(fb_net, vertex.size=my_friends$degree/20, vertex.label=NA,
vertex.color=my_friends$color, edge.color=my_network_el$color,
edge.width=1, edge.arrow.size=0, edge.curved=0.3, layout=l)
## 3.2. 프로필 사진 넣어 시각화
# 레이아웃 -1에서 1로 척도 재조정
l <- layout.norm(l, -1, 1, -1, 1)
plot(fb_net, vertex.size=4, vertex.label=NA, edge.color=my_network_el$color,
vertex.shape="square",vertex.color="white", vertex.frame.color="white",
edge.width=1.5, edge.arrow.size=0, edge.curved=0.2, layout=l)
img.sc <- 0.05 #Image scaling
for (i in 1:dim(l)[1]) {
img <- my_friends$picture.dld[i]
img <- if(grepl(".jpg", img)) readJPEG(img) else "gray20"
rasterImage(img, l[i,1]-img.sc, l[i,2]-img.sc, l[i,1]+img.sc, l[i,2]+img.sc)
}