List of titles for WoS islands

Second version (September 2013)

WoS2Pajek 1.2 (September 2013) creates also a csv file titles.csv with works names and their title, first author, journal and year of publication. This file can be merged with a subnetwork (island) net file to produce a list of titles of works contained in the subnetwork. It uses the Python csv library.

> setwd("E:/Data/Centrality/net")
> Code <- function(x) paste(substr(unlist(strsplit(x," ")),1,1),collapse="")
> T <- read.csv('../WoS/titles.csv',sep=";",colClasses="character")
> dim(T)
[1] 38535     6
> names(T)
[1] "name"    "WoSline" "author"  "title"   "journal" "year"   
> T$WoSline <- as.integer(T$WoSline)
> T$year <- as.integer(T$year)
> n <- nrow(T)
> cod <- character(n)
> for(i in 1:n) cod[i] <- Code(as.character(T$journal[i]))
> T$code <- cod
> save(T,file="titlesCentr.Rdata")
> dim(T)
[1] 38535     7
#
# description produces a csv file csvFile containing a table with columns
#   (name, WoSline, author, title, journal, year, code) 
# for vertices of a subnetwork from file netFile.
# The titles for all (DC>0) vertices are stored in the table T.
# Since for vertices with DC=0 the title is not available it takes '***'.
#
# Vladimir Batagelj,  September 7, 2013 / August 19, 2013
#
description <- function(netFile,csvFile,T){
  f <- function(L){
    s <- unlist(strsplit(L,"[[:space:]]+"))
    if(s[1]=="") t <- s[3] else t <- s[2]
    t <- substr(t,2,nchar(t)-1)
    if(substr(t,1,1)=="=") t <- substr(t,2,nchar(t))
    return(t)
  }
  net <- file(netFile,"r")
  L <- readLines(net,n=1)       
  n <- as.integer(unlist(strsplit(L,"[[:space:]]+"))[2])
  S <- readLines(net,n=n);  close(net)
  N <- gsub("&#39;","'",unlist(lapply(S,f)))
  nn <- length(N); tit <- character(nn); aut <- character(nn); jrn <- character(nn);
  cod <- character(nn); wos <- integer(nn); yer <- integer(nn)
  for(k in 1:nn){j <- which(T$name==N[k]); 
    tit[k] <- ifelse(length(j)==0,NA,T$title[j])
    aut[k] <- ifelse(length(j)==0,NA,T$author[j])
    wos[k] <- ifelse(length(j)==0,NA,T$WoSline[j])
    jrn[k] <- ifelse(length(j)==0,NA,T$journal[j])
    cod[k] <- ifelse(length(j)==0,NA,T$code[j])
    yer[k] <- ifelse(length(j)==0,NA,T$year[j])
    if(is.na(yer[k])){
      l <- regexpr("(",N[k],fixed=TRUE)[1]+1
      r <- regexpr(")",N[k],fixed=TRUE)[1]-1
      yer[k] <- as.integer(substr(N[k],l,r))
    }
  }
  D <- data.frame(name=as.character(N),WoSline=wos,author=as.character(aut),
     title=as.character(tit),journal=as.character(jrn),year=yer,code=as.character(cod))
  write.csv2(D,file=csvFile,fileEncoding="UTF-8",row.names=FALSE,quote=FALSE)
  return(D)
}

Application:

> setwd("E:/Data/Centrality/net")
> load("titlesCentr.Rdata")
> names(T)
[1] "name"    "WoSline" "author"  "title"   "journal" "year"    "code"   
> dim(T)
[1] 38535     7
> source("E:\\Data\\Centrality\\net\\res\\description.R")
> d <- description("./res/island2.net","./res/island2new.csv",T)

First version (August 2013)

WoS2Pajek 1.2 (August 2013) creates also a csv file titles.csv with works names and their titles. This file can be merged with a subnetwork (island) net file to produce a list of titles of works contained in the subnetwork.

> setwd("D:/Data/Centrality/WoS")
> T <- read.csv("titles.csv",header=FALSE,sep="÷",quote="",colClasses="character")
> t <- T$V1[1]
> s <- substr(t,23,23)
> U <- as.data.frame(matrix(unlist(strsplit(T$V1,s)),ncol=2,byrow=TRUE))
> names(U) <- c("name","title")
> library(gdata)
> U$title <- trim(U$title)
> T$name <- as.character(U$name)
> T$title <- as.character(U$title)
> write.table(T,file="CentTitles.csv",sep=";",qmethod="double",row.names=FALSE)
> save(T,file="CentTitles.Rdata")

> write.table(T,file="CentTit.csv",sep=";",fileEncoding="UTF-8",qmethod="double",row.names=FALSE,quote=FALSE) 

Function in R to create a list:

#
# description produces a csv file csvFile containing a table with columns
# (name, year, title) for vertices of a subnetwork from file netFile.
# The titles for all (DC>0) vertices are stored in the table T.
# Since for vertices with DC=0 the title is not available it takes '***'.
#
# Vladimir Batagelj,  August 19, 2013
#
description <- function(netFile,csvFile,T){
  f <- function(L){
    s <- unlist(strsplit(L,"[[:space:]]+"))
    if(s[1]=="") t <- s[3] else t <- s[2]
    t <- substr(t,2,nchar(t)-1)
    if(substr(t,1,1)=="=") t <- substr(t,2,nchar(t))
    return(t)
  }
  getYear <- function(x){
    i <- regexpr('\\(',x)[1]+1; j <- regexpr('\\)',x)[1]-1
    ifelse(i<j,as.integer(substr(x,i,j)),-1)
  }
  net <- file(netFile,"r")
  L <- readLines(net,n=1)       
  n <- as.integer(unlist(strsplit(L,"[[:space:]]+"))[2])
  S <- readLines(net,n=n);  close(net)
  N <- gsub("&#39;","'",unlist(lapply(S,f)))
  year <- unlist(lapply(N,getYear))
  nn <- length(N); tit <- character(nn)
  for(k in 1:nn){j <- which(T$name==N[k]); tit[k] <- ifelse(length(j)==0,"***",T$title[j])}
  D <- data.frame(name=as.character(N),year=year,title=as.character(tit))
  write.table(D,file=csvFile,sep=";",fileEncoding="UTF-8",qmethod="double",row.names=FALSE)
  return(D)
}

# setwd("D:/Data/Centrality/net")
# source("D:\\Data\\Centrality\\net\\res\\description.R")
# T <- read.csv("CentTitles.csv",header=TRUE,sep=";",colClasses="character")
# d <- description("mainPath.net","mainPath.csv",T)
# d <- description("islands2.net","islands2.csv",T)

Application:

setwd("D:/Data/Centrality/net")
source("D:\\Data\\Centrality\\net\\res\\description.R")
T <- read.csv("../WoS/CentTitles.csv",header=TRUE,sep=";",colClasses="character")
d <- description("./res/mainPath.net","./res/mainPath.csv",T)

Instead of the command T ← read.csv we can use also

T <- load("CentTitles.Rdata")
notes/net/titls.txt · Last modified: 2015/07/16 22:13 by vlado
 
Except where otherwise noted, content on this wiki is licensed under the following license: CC Attribution-Noncommercial-Share Alike 3.0 Unported
Recent changes RSS feed Donate Powered by PHP Valid XHTML 1.0 Valid CSS Driven by DokuWiki