mxmehl/uni-ba-socialagenda
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

import issue csv-files into issues and tweets DF

Browse Source
This commit is contained in:
mxmehl
2015-02-24 18:19:46 +01:00
parent f9757f1a0a
commit 6af9dfb513
3 changed files with 281 additions and 500 deletions
Download Patch File Download Diff File Expand all files Collapse all files
.Rhistory
+216 -486
View File
@@ -1,93 +1,3 @@
require(stringr)
library(foreach)
library(doParallel)
# MATCH TWEETS ------------------------------------------------------------
id_folder <- "matched-ids"
unlink(id_folder, recursive = TRUE)
dir.create(id_folder)
issues <- data.frame(date = drange)
issuelist <- readLines("issues.xml")
issuelist <- str_replace_all(string = issuelist, pattern = ".*<!-- .+ -->", "")
issuelist <- xmlToList(issuelist)
issueheads <- names(issuelist)
issues[issueheads] <- 0
tweets$issue <- ""
tweets$tags <- ""
tagexpand <- c("", "s", "n", "en", "er", "e")
# Parallelisation
writeLines(c(""), "issuecomp-analysis.log")
cl<-makeCluster(4)
registerDoParallel(cl)
foreach(d = 1:nrow(issues), .packages = c("stringr"), .combine=rbind) %dopar% {
#for(d in 1:nrow(issues)) {
# Go through every day
curdate <- issues$date[d]
cat(paste(as.character(curdate),"\n"), file="issuecomp-analysis.log", append=TRUE)
# Put all tweets from specific day in a temporary DF
tweets_curday <- tweets[tweets[, "created_at"] == curdate, ]
for(t in 1:nrow(tweets_curday)){
# cat(paste("Starting tweet", t, "of",as.character(curdate),"\n"), file="issuecomp-analysis.log", append=TRUE)
# Select tweet's text, make it lowercase and remove hashtag indicators (#)
curtext <- as.character(tweets_curday$text[t])
curtext <- str_replace_all(curtext, "#", "")
curid <- as.character(tweets_curday$id_str[t])
# Now test each single issue (not tag!)
for(i in 1:length(issueheads)) {
curissue <- issueheads[i]
curtags <- as.character(issuelist[[curissue]])
curfile <- str_c(id_folder,"/",curissue,".csv")
# Now test all tags of a single issue
for(s in 1:length(curtags)) {
curtag <- curtags[s]
curchars <- nchar(curtag, type = "chars")
# Check if tag is an acronym. If so, ignore.case will be deactivated in smartPatternMatch
if(curchars <= 4) {
curacro <- checkAcronym(string = curtag, chars = curchars)
} else {
curacro <- FALSE
}
# Now expand the current tag by possible suffixes that may be plural forms
if(!curacro) {
for(e in 1:length(tagexpand)) {
curtag[e] <- str_c(curtag[1], tagexpand[e])
}
}
# Set Levenshtein distance depending on char length
if(curchars <= 4) {
curdistance <- 0
} else {
curdistance <- 1
}
# Match current tweet with tag. If >= 5 letters allow 1 changed letter, if >=8 letters allow also 1 (Levenshtein distance)
tags_found <- NULL
# Match the tweet with each variation of tagexpand
for(e in 1:length(curtag)) {
tags_found[e] <- smartPatternMatch(curtext, curtag[e], curdistance, curacro)
}
tags_found <- any(tags_found)
curtag <- curtag[1]
if(tags_found == TRUE) {
# # Raise number of findings on this day for this issue by 1
# issues[d,curissue] <- issues[d,curissue] + 1
#
# # Add issue and first matched tag of tweet to tweets-DF
# oldissue <- tweets[tweets[, "id_str"] == curid, "issue"]
# tweets[tweets[, "id_str"] == curid, "issue"] <- str_c(oldissue, curissue, ";")
# oldtag <- tweets[tweets[, "id_str"] == curid, "tags"]
# tweets[tweets[, "id_str"] == curid, "tags"] <- str_c(oldtag, curtag, ";")
# Add information to file for function viewPatternMatching
write(str_c(curdate,";\"",curid,"\";",curissue,";",curtag), curfile, append = TRUE)
cat(paste("Match!\n"), file="issuecomp-analysis.log", append=TRUE)
# data.frame(date=curdate, issue=curissue)
break # next issue, no more tags from same issue
}
else {
#cat("Nothing found\n")
}
} # /for curtags
} # /for issuelist
} # /for tweets_curday
} # /for drange
require(lubridate) require(lubridate)
require(XML) require(XML)
require(ggplot2) require(ggplot2)
@@ -95,10 +5,19 @@ require(reshape2)
require(stringr) require(stringr)
library(foreach) library(foreach)
library(doParallel) library(doParallel)
# MATCH TWEETS ------------------------------------------------------------ source("issuecomp-functions.R")
id_folder <- "matched-ids" getwd()
unlink(id_folder, recursive = TRUE) setwd("~/Dokumente/Uni/Aktuell/BA-Arbeit/uni-ba-issuecomp")
dir.create(id_folder) getwd()
list.files()
list.files("matched-ids/")
load(file = "tweets_untagged.RData")
issues <- data.frame(date = drange)
# Create date range
date_start <- as.Date("2014-01-01")
date_end <- as.Date("2014-12-31")
drange <- as.integer(date_end - date_start)
drange <- date_start + days(0:drange)
issues <- data.frame(date = drange) issues <- data.frame(date = drange)
issuelist <- readLines("issues.xml") issuelist <- readLines("issues.xml")
issuelist <- str_replace_all(string = issuelist, pattern = ".*<!-- .+ -->", "") issuelist <- str_replace_all(string = issuelist, pattern = ".*<!-- .+ -->", "")
@@ -107,86 +26,148 @@ issueheads <- names(issuelist)
issues[issueheads] <- 0 issues[issueheads] <- 0
tweets$issue <- "" tweets$issue <- ""
tweets$tags <- "" tweets$tags <- ""
tagexpand <- c("", "s", "n", "en", "er", "e")
# Parallelisation
writeLines(c(""), "issuecomp-analysis.log")
cl<-makeCluster(4)
registerDoParallel(cl)
foreach(d = 1:nrow(issues), .packages = c("stringr"), .combine=rbind) %dopar% {
#for(d in 1:nrow(issues)) {
# Go through every day
curdate <- issues$date[d]
cat(paste(as.character(curdate),"\n"), file="issuecomp-analysis.log", append=TRUE)
# Put all tweets from specific day in a temporary DF
tweets_curday <- tweets[tweets[, "created_at"] == curdate, ]
for(t in 1:nrow(tweets_curday)){
# cat(paste("Starting tweet", t, "of",as.character(curdate),"\n"), file="issuecomp-analysis.log", append=TRUE)
# Select tweet's text, make it lowercase and remove hashtag indicators (#)
curtext <- as.character(tweets_curday$text[t])
curtext <- str_replace_all(curtext, "#", "")
curid <- as.character(tweets_curday$id_str[t])
# Now test each single issue (not tag!)
for(i in 1:length(issueheads)) {
curissue <- issueheads[i]
curtags <- as.character(issuelist[[curissue]])
curfile <- str_c(id_folder,"/",curissue,".csv")
# Now test all tags of a single issue
for(s in 1:length(curtags)) {
curtag <- curtags[s]
curchars <- nchar(curtag, type = "chars")
# Check if tag is an acronym. If so, ignore.case will be deactivated in smartPatternMatch
if(curchars <= 4) {
curacro <- checkAcronym(string = curtag, chars = curchars)
} else {
curacro <- FALSE
}
# Now expand the current tag by possible suffixes that may be plural forms
if(!curacro) {
for(e in 1:length(tagexpand)) {
curtag[e] <- str_c(curtag[1], tagexpand[e])
}
}
# Set Levenshtein distance depending on char length
if(curchars <= 4) {
curdistance <- 0
} else {
curdistance <- 1
}
# Match current tweet with tag. If >= 5 letters allow 1 changed letter, if >=8 letters allow also 1 (Levenshtein distance)
tags_found <- NULL
# Match the tweet with each variation of tagexpand
for(e in 1:length(curtag)) {
tags_found[e] <- smartPatternMatch(curtext, curtag[e], curdistance, curacro)
}
tags_found <- any(tags_found)
curtag <- curtag[1]
if(tags_found == TRUE) {
# # Raise number of findings on this day for this issue by 1
# issues[d,curissue] <- issues[d,curissue] + 1
#
# # Add issue and first matched tag of tweet to tweets-DF
# oldissue <- tweets[tweets[, "id_str"] == curid, "issue"]
# tweets[tweets[, "id_str"] == curid, "issue"] <- str_c(oldissue, curissue, ";")
# oldtag <- tweets[tweets[, "id_str"] == curid, "tags"]
# tweets[tweets[, "id_str"] == curid, "tags"] <- str_c(oldtag, curtag, ";")
# Add information to file for function viewPatternMatching
write(str_c(curdate,";\"",curid,"\";",curissue,";",curtag), curfile, append = TRUE)
# cat(paste("Match!\n"), file="issuecomp-analysis.log", append=TRUE)
# data.frame(date=curdate, issue=curissue)
break # next issue, no more tags from same issue
}
else {
#cat("Nothing found\n")
}
} # /for curtags
} # /for issuelist
} # /for tweets_curday
} # /for drange
stopCluster(cl)
drange
drange[40]
drange[50]
View(issues) View(issues)
list.files("matched-ids/")
results <- list.files("matched-ids/")
results
read.csv("matched-ids/i10.trans.csv")
read.csv("matched-ids/i10.trans.csv", sep=";")
read.csv("matched-ids/i10.trans.csv", sep=";", stringsAsFactors=F)
read.csv("matched-ids/i10.trans.csv", sep=";", stringsAsFactors=T)
reesult_files <- read.csv("matched-ids/i10.trans.csv", sep=";", stringsAsFactors=F)
View(reesult_files)
result_files <- read.csv("matched-ids/i10.trans.csv", sep=";", colClasses=c("date", "character", "character", "character"))
result_files <- read.csv("matched-ids/i10.trans.csv", sep=";", colClasses=c("character", "character", "character", "character"))
rm(reesult_files)
View(result_files)
nrow(result_files)
result_files <- result_files(!duplicated(result_files))
result_files <- result_files[!duplicated(result_files)]
result_files <- result_files[!duplicated(result_files), ]
nrow(result_files)
result_files <- read.csv("matched-ids/i10.trans.csv", sep=";", colClasses=c("character", "character", "character", "character"), header=F)
View(result_files)
read.results
results
setwd("matched-ids/")
list.files("")
getwd()
list.files()
results <- list.files()
results
results_cat <- read.csv(results, sep=";", colClasses=c("character", "character", "character", "character"), header=F)
results_cat <- read.csv(results[1], sep=";", colClasses=c("character", "character", "character", "character"), header=F)
results_cat
View(results_cat)
source("issuecomp-functions.R")
setwd("~/Dokumente/Uni/Aktuell/BA-Arbeit/uni-ba-issuecomp")
source("issuecomp-functions.R")
insertRow
results_temp <- read.csv(results[2], sep=";", colClasses=c("character", "character", "character", "character"), header=F)
setwd("matched-ids/")
results_temp <- read.csv(results[2], sep=";", colClasses=c("character", "character", "character", "character"), header=F)
rm(result_files)
insertRow(existingDF = results_cat, results_temp)
rm(results_cat)
for(r in 1:length(results)) {
if(r == 1) {
results_cat <- read.csv(results[r], sep=";", colClasses=c("character", "character", "character", "character"), header=F)
} else {
results_temp <- read.csv(results[r], sep=";", colClasses=c("character", "character", "character", "character"), header=F)
insertRow(results_cat, results_temp)
}
}
for(r in 1:length(results)) {
if(r == 1) {
results_cat <- read.csv(results[r], sep=";", colClasses=c("character", "character", "character", "character"), header=F)
} else {
results_temp <- read.csv(results[r], sep=";", colClasses=c("character", "character", "character", "character"), header=F)
results_cat insertRow(results_cat, results_temp)
}
}
for(r in 1:length(results)) {
if(r == 1) {
results_cat <- read.csv(results[r], sep=";", colClasses=c("character", "character", "character", "character"), header=F)
} else {
results_temp <- read.csv(results[r], sep=";", colClasses=c("character", "character", "character", "character"), header=F)
results_cat <- insertRow(results_cat, results_temp)
}
}
View(results_cat)
results_cat[20000]
results_cat[20000, ]
rm(r, results_temp)
results_cat <- results_cat[!duplicated(results_cat), ]
View(results_cat)
rm(results, results_cat)
results_files <- list.files()
for(r in 1:length(results)) {
if(r == 1) {
results <- read.csv(results_files[r], sep=";", colClasses=c("character", "character", "character", "character"), header=F)
} else {
results_temp <- read.csv(results_files[r], sep=";", colClasses=c("character", "character", "character", "character"), header=F)
results <- insertRow(results_cat, results_temp)
}
}
rm(r, results_temp)
results <- results[!duplicated(results), ]
results_files <- list.files()
for(r in 1:length(results_files)) {
if(r == 1) {
results <- read.csv(results_files[r], sep=";", colClasses=c("character", "character", "character", "character"), header=F)
} else {
results_temp <- read.csv(results_files[r], sep=";", colClasses=c("character", "character", "character", "character"), header=F)
results <- insertRow(results, results_temp)
}
}
rm(r, results_temp)
results <- results[!duplicated(results), ]
View(results)
View(issues)
row.names(results) <- NULL
View(results)
rownames(results)
row.names(results)
names(results)
View(tweets)
View(tweets)
names(results) <- c("date", "id_str", "issue", "tags")
View(results)
results_test <- results[order(results$id_str)]
results_test <- results[order(results$id_str), ]
View(results_test)
results_files <- list.files()
for(r in 1:length(results_files)) {
if(r == 1) {
results <- read.csv(results_files[r], sep=";", colClasses=c("character", "character", "character", "character"), header=F)
} else {
results_temp <- read.csv(results_files[r], sep=";", colClasses=c("character", "character", "character", "character"), header=F)
results <- insertRow(results, results_temp)
}
}
rm(r, results_temp)
rm(r, results_temp, results_files)
results <- results[!duplicated(results), ]
names(results)
names(results) <- c("date", "id_str", "issue", "tags")
View(results)
results_test <- results[order(results$id_str), ]
row.names(results) <- NULL
results <- results[order(results$id_str), ]
row.names(results) <- NULL
View(results)
rm(results_test)
View(issues)
as.character(results$date[2])
class(results$date)
class(issues$date)
View(issues)
as.character(issues$date[2])
issues$date[2]
issuelist <- readLines("issues.xml")
issuelist <- str_replace_all(string = issuelist, pattern = ".*<!-- .+ -->", "")
issuelist <- xmlToList(issuelist)
issueheads <- names(issuelist)
require(lubridate) require(lubridate)
require(XML) require(XML)
require(ggplot2) require(ggplot2)
@@ -194,319 +175,68 @@ require(reshape2)
require(stringr) require(stringr)
library(foreach) library(foreach)
library(doParallel) library(doParallel)
drange[70] issuelist <- readLines("issues.xml")
drange[80] issuelist <- str_replace_all(string = issuelist, pattern = ".*<!-- .+ -->", "")
drange[90] issuelist <- xmlToList(issuelist)
cl<-makeCluster(4) issueheads <- names(issuelist)
registerDoParallel(cl) setwd("~/Dokumente/Uni/Aktuell/BA-Arbeit/uni-ba-issuecomp")
foreach(d = 51:90, .packages = c("stringr"), .combine=rbind) %dopar% { issuelist <- readLines("issues.xml")
#for(d in 1:nrow(issues)) { issuelist <- str_replace_all(string = issuelist, pattern = ".*<!-- .+ -->", "")
# Go through every day issuelist <- xmlToList(issuelist)
curdate <- issues$date[d] issueheads <- names(issuelist)
cat(paste(as.character(curdate),"\n"), file="issuecomp-analysis.log", append=TRUE) issues[issueheads] <- 0
# Put all tweets from specific day in a temporary DF curdate <- as.character(results$date[3])
tweets_curday <- tweets[tweets[, "created_at"] == curdate, ] curissue <- as.character(results$issue[3])
for(t in 1:nrow(tweets_curday)){ curdate
# cat(paste("Starting tweet", t, "of",as.character(curdate),"\n"), file="issuecomp-analysis.log", append=TRUE) curissue
# Select tweet's text, make it lowercase and remove hashtag indicators (#) issues[curdate, curissue] <- issues[curdate, curissue] + 1
curtext <- as.character(tweets_curday$text[t]) View(issues)
curtext <- str_replace_all(curtext, "#", "") issues <- data.frame(date = drange)
curid <- as.character(tweets_curday$id_str[t]) issues[issueheads] <- 0
# Now test each single issue (not tag!) View(issues)
for(i in 1:length(issueheads)) { issues[issues[, "date"] == curdate, curissue] <- issues[issues[, "date"] == curdate, curissue] + 1
curissue <- issueheads[i] View(issues)
curtags <- as.character(issuelist[[curissue]]) for(r in 1:nrow(results)) {
curfile <- str_c(id_folder,"/",curissue,".csv") curdate <- as.character(results$date[r])
# Now test all tags of a single issue curissue <- as.character(results$issue[r])
for(s in 1:length(curtags)) { issues[issues[, "date"] == curdate, curissue] <- issues[issues[, "date"] == curdate, curissue] + 1
curtag <- curtags[s]
curchars <- nchar(curtag, type = "chars")
# Check if tag is an acronym. If so, ignore.case will be deactivated in smartPatternMatch
if(curchars <= 4) {
curacro <- checkAcronym(string = curtag, chars = curchars)
} else {
curacro <- FALSE
} }
# Now expand the current tag by possible suffixes that may be plural forms View(issues)
if(!curacro) { issues[issueheads] <- 0
for(e in 1:length(tagexpand)) { View(issues)
curtag[e] <- str_c(curtag[1], tagexpand[e]) for(r in 1:nrow(results)) {
curdate <- as.character(results$date[r])
curid <- as.character(results$id_str[r])
curissue <- as.character(results$issue[r])
curtag <- as.character(results$tags[r])
# Update issue counter (date and issue)
issues[issues[, "date"] == curdate, curissue] <- issues[issues[, "date"] == curdate, curissue] + 1
# Update tweet dataframe (id, issue and tags)
oldissue <- tweets[tweets[, "id_str"] == curid, "issue"]
tweets[tweets[, "id_str"] == curid, "issue"] <- str_c(oldissue, curissue, ",")
oldtag <- tweets[tweets[, "id_str"] == curid, "tags"]
tweets[tweets[, "id_str"] == curid, "tags"] <- str_c(oldtag, curtag, ",")
} }
View(tweets)
tweets$issue <- ""
tweets$tags <- ""
View(tweets)
issues[issueheads] <- 0
for(r in 1:nrow(results)) {
curdate <- as.character(results$date[r])
curid <- as.character(results$id_str[r])
curissue <- as.character(results$issue[r])
curtag <- as.character(results$tags[r])
cat("Sorting match", r, "from", nrow(results), "\n")
# Update issue counter (date and issue)
issues[issues[, "date"] == curdate, curissue] <- issues[issues[, "date"] == curdate, curissue] + 1
# Update tweet dataframe (id, issue and tags)
oldissue <- tweets[tweets[, "id_str"] == curid, "issue"]
tweets[tweets[, "id_str"] == curid, "issue"] <- str_c(oldissue, curissue, ",")
oldtag <- tweets[tweets[, "id_str"] == curid, "tags"]
tweets[tweets[, "id_str"] == curid, "tags"] <- str_c(oldtag, curtag, ",")
} }
# Set Levenshtein distance depending on char length View(issues)
if(curchars <= 4) { View(tweets)
curdistance <- 0 View(tweets)
} else { save(tweets, file="tweets_tagged.RData")
curdistance <- 1
}
# Match current tweet with tag. If >= 5 letters allow 1 changed letter, if >=8 letters allow also 1 (Levenshtein distance)
tags_found <- NULL
# Match the tweet with each variation of tagexpand
for(e in 1:length(curtag)) {
tags_found[e] <- smartPatternMatch(curtext, curtag[e], curdistance, curacro)
}
tags_found <- any(tags_found)
curtag <- curtag[1]
if(tags_found == TRUE) {
# # Raise number of findings on this day for this issue by 1
# issues[d,curissue] <- issues[d,curissue] + 1
#
# # Add issue and first matched tag of tweet to tweets-DF
# oldissue <- tweets[tweets[, "id_str"] == curid, "issue"]
# tweets[tweets[, "id_str"] == curid, "issue"] <- str_c(oldissue, curissue, ";")
# oldtag <- tweets[tweets[, "id_str"] == curid, "tags"]
# tweets[tweets[, "id_str"] == curid, "tags"] <- str_c(oldtag, curtag, ";")
# Add information to file for function viewPatternMatching
write(str_c(curdate,";\"",curid,"\";",curissue,";",curtag), curfile, append = TRUE)
# cat(paste("Match!\n"), file="issuecomp-analysis.log", append=TRUE)
# data.frame(date=curdate, issue=curissue)
break # next issue, no more tags from same issue
}
else {
#cat("Nothing found\n")
}
} # /for curtags
} # /for issuelist
} # /for tweets_curday
} # /for drange
#rm(tweets_curday,curacro, curchars, curdate,curfile,curid,curissue,curtag,curtags,curtext,d,date_end,date_start,i,id_folder,oldissue,oldtag,s,t,tags_found)
stopCluster(cl)
drange[121]
cl<-makeCluster(4)
registerDoParallel(cl)
foreach(d = 91:120, .packages = c("stringr"), .combine=rbind) %dopar% {
#for(d in 1:nrow(issues)) {
# Go through every day
curdate <- issues$date[d]
cat(paste(as.character(curdate),"\n"), file="issuecomp-analysis.log", append=TRUE)
# Put all tweets from specific day in a temporary DF
tweets_curday <- tweets[tweets[, "created_at"] == curdate, ]
for(t in 1:nrow(tweets_curday)){
# cat(paste("Starting tweet", t, "of",as.character(curdate),"\n"), file="issuecomp-analysis.log", append=TRUE)
# Select tweet's text, make it lowercase and remove hashtag indicators (#)
curtext <- as.character(tweets_curday$text[t])
curtext <- str_replace_all(curtext, "#", "")
curid <- as.character(tweets_curday$id_str[t])
# Now test each single issue (not tag!)
for(i in 1:length(issueheads)) {
curissue <- issueheads[i]
curtags <- as.character(issuelist[[curissue]])
curfile <- str_c(id_folder,"/",curissue,".csv")
# Now test all tags of a single issue
for(s in 1:length(curtags)) {
curtag <- curtags[s]
curchars <- nchar(curtag, type = "chars")
# Check if tag is an acronym. If so, ignore.case will be deactivated in smartPatternMatch
if(curchars <= 4) {
curacro <- checkAcronym(string = curtag, chars = curchars)
} else {
curacro <- FALSE
}
# Now expand the current tag by possible suffixes that may be plural forms
if(!curacro) {
for(e in 1:length(tagexpand)) {
curtag[e] <- str_c(curtag[1], tagexpand[e])
}
}
# Set Levenshtein distance depending on char length
if(curchars <= 4) {
curdistance <- 0
} else {
curdistance <- 1
}
# Match current tweet with tag. If >= 5 letters allow 1 changed letter, if >=8 letters allow also 1 (Levenshtein distance)
tags_found <- NULL
# Match the tweet with each variation of tagexpand
for(e in 1:length(curtag)) {
tags_found[e] <- smartPatternMatch(curtext, curtag[e], curdistance, curacro)
}
tags_found <- any(tags_found)
curtag <- curtag[1]
if(tags_found == TRUE) {
# # Raise number of findings on this day for this issue by 1
# issues[d,curissue] <- issues[d,curissue] + 1
#
# # Add issue and first matched tag of tweet to tweets-DF
# oldissue <- tweets[tweets[, "id_str"] == curid, "issue"]
# tweets[tweets[, "id_str"] == curid, "issue"] <- str_c(oldissue, curissue, ";")
# oldtag <- tweets[tweets[, "id_str"] == curid, "tags"]
# tweets[tweets[, "id_str"] == curid, "tags"] <- str_c(oldtag, curtag, ";")
# Add information to file for function viewPatternMatching
write(str_c(curdate,";\"",curid,"\";",curissue,";",curtag), curfile, append = TRUE)
# cat(paste("Match!\n"), file="issuecomp-analysis.log", append=TRUE)
# data.frame(date=curdate, issue=curissue)
break # next issue, no more tags from same issue
}
else {
#cat("Nothing found\n")
}
} # /for curtags
} # /for issuelist
} # /for tweets_curday
} # /for drange
stopCluster(cl)
drange[102]
require(lubridate)
require(XML)
require(ggplot2)
require(reshape2)
require(stringr)
library(foreach)
library(doParallel)
cl<-makeCluster(4)
registerDoParallel(cl)
foreach(d = 101:nrow(issues), .packages = c("stringr"), .combine=rbind) %dopar% {
#for(d in 1:nrow(issues)) {
# Go through every day
curdate <- issues$date[d]
cat(paste(as.character(curdate),"\n"), file="issuecomp-analysis.log", append=TRUE)
# Put all tweets from specific day in a temporary DF
tweets_curday <- tweets[tweets[, "created_at"] == curdate, ]
for(t in 1:nrow(tweets_curday)){
# cat(paste("Starting tweet", t, "of",as.character(curdate),"\n"), file="issuecomp-analysis.log", append=TRUE)
# Select tweet's text, make it lowercase and remove hashtag indicators (#)
curtext <- as.character(tweets_curday$text[t])
curtext <- str_replace_all(curtext, "#", "")
curid <- as.character(tweets_curday$id_str[t])
# Now test each single issue (not tag!)
for(i in 1:length(issueheads)) {
curissue <- issueheads[i]
curtags <- as.character(issuelist[[curissue]])
curfile <- str_c(id_folder,"/",curissue,".csv")
# Now test all tags of a single issue
for(s in 1:length(curtags)) {
curtag <- curtags[s]
curchars <- nchar(curtag, type = "chars")
# Check if tag is an acronym. If so, ignore.case will be deactivated in smartPatternMatch
if(curchars <= 4) {
curacro <- checkAcronym(string = curtag, chars = curchars)
} else {
curacro <- FALSE
}
# Now expand the current tag by possible suffixes that may be plural forms
if(!curacro) {
for(e in 1:length(tagexpand)) {
curtag[e] <- str_c(curtag[1], tagexpand[e])
}
}
# Set Levenshtein distance depending on char length
if(curchars <= 4) {
curdistance <- 0
} else {
curdistance <- 1
}
# Match current tweet with tag. If >= 5 letters allow 1 changed letter, if >=8 letters allow also 1 (Levenshtein distance)
tags_found <- NULL
# Match the tweet with each variation of tagexpand
for(e in 1:length(curtag)) {
tags_found[e] <- smartPatternMatch(curtext, curtag[e], curdistance, curacro)
}
tags_found <- any(tags_found)
curtag <- curtag[1]
if(tags_found == TRUE) {
# # Raise number of findings on this day for this issue by 1
# issues[d,curissue] <- issues[d,curissue] + 1
#
# # Add issue and first matched tag of tweet to tweets-DF
# oldissue <- tweets[tweets[, "id_str"] == curid, "issue"]
# tweets[tweets[, "id_str"] == curid, "issue"] <- str_c(oldissue, curissue, ";")
# oldtag <- tweets[tweets[, "id_str"] == curid, "tags"]
# tweets[tweets[, "id_str"] == curid, "tags"] <- str_c(oldtag, curtag, ";")
# Add information to file for function viewPatternMatching
write(str_c(curdate,";\"",curid,"\";",curissue,";",curtag), curfile, append = TRUE)
# cat(paste("Match!\n"), file="issuecomp-analysis.log", append=TRUE)
# data.frame(date=curdate, issue=curissue)
break # next issue, no more tags from same issue
}
else {
#cat("Nothing found\n")
}
} # /for curtags
} # /for issuelist
} # /for tweets_curday
} # /for drange
require(lubridate)
require(XML)
require(ggplot2)
require(reshape2)
require(stringr)
library(foreach)
library(doParallel)
cl<-makeCluster(3)
registerDoParallel(cl)
foreach(d = 101:nrow(issues), .packages = c("stringr"), .combine=rbind) %dopar% {
#for(d in 1:nrow(issues)) {
# Go through every day
curdate <- issues$date[d]
cat(paste(as.character(curdate),"\n"), file="issuecomp-analysis.log", append=TRUE)
# Put all tweets from specific day in a temporary DF
tweets_curday <- tweets[tweets[, "created_at"] == curdate, ]
for(t in 1:nrow(tweets_curday)){
# cat(paste("Starting tweet", t, "of",as.character(curdate),"\n"), file="issuecomp-analysis.log", append=TRUE)
# Select tweet's text, make it lowercase and remove hashtag indicators (#)
curtext <- as.character(tweets_curday$text[t])
curtext <- str_replace_all(curtext, "#", "")
curid <- as.character(tweets_curday$id_str[t])
# Now test each single issue (not tag!)
for(i in 1:length(issueheads)) {
curissue <- issueheads[i]
curtags <- as.character(issuelist[[curissue]])
curfile <- str_c(id_folder,"/",curissue,".csv")
# Now test all tags of a single issue
for(s in 1:length(curtags)) {
curtag <- curtags[s]
curchars <- nchar(curtag, type = "chars")
# Check if tag is an acronym. If so, ignore.case will be deactivated in smartPatternMatch
if(curchars <= 4) {
curacro <- checkAcronym(string = curtag, chars = curchars)
} else {
curacro <- FALSE
}
# Now expand the current tag by possible suffixes that may be plural forms
if(!curacro) {
for(e in 1:length(tagexpand)) {
curtag[e] <- str_c(curtag[1], tagexpand[e])
}
}
# Set Levenshtein distance depending on char length
if(curchars <= 4) {
curdistance <- 0
} else {
curdistance <- 1
}
# Match current tweet with tag. If >= 5 letters allow 1 changed letter, if >=8 letters allow also 1 (Levenshtein distance)
tags_found <- NULL
# Match the tweet with each variation of tagexpand
for(e in 1:length(curtag)) {
tags_found[e] <- smartPatternMatch(curtext, curtag[e], curdistance, curacro)
}
tags_found <- any(tags_found)
curtag <- curtag[1]
if(tags_found == TRUE) {
# # Raise number of findings on this day for this issue by 1
# issues[d,curissue] <- issues[d,curissue] + 1
#
# # Add issue and first matched tag of tweet to tweets-DF
# oldissue <- tweets[tweets[, "id_str"] == curid, "issue"]
# tweets[tweets[, "id_str"] == curid, "issue"] <- str_c(oldissue, curissue, ";")
# oldtag <- tweets[tweets[, "id_str"] == curid, "tags"]
# tweets[tweets[, "id_str"] == curid, "tags"] <- str_c(oldtag, curtag, ";")
# Add information to file for function viewPatternMatching
write(str_c(curdate,";\"",curid,"\";",curissue,";",curtag), curfile, append = TRUE)
# cat(paste("Match!\n"), file="issuecomp-analysis.log", append=TRUE)
# data.frame(date=curdate, issue=curissue)
break # next issue, no more tags from same issue
}
else {
#cat("Nothing found\n")
}
} # /for curtags
} # /for issuelist
} # /for tweets_curday
} # /for drange
stopCluster(cl)
drange[200]
drange[300]
drange[280]
drange[270]
drange[259]
issuecomp-analysis.R
+65 -14
View File
@@ -7,6 +7,7 @@ library(foreach)
library(doParallel) library(doParallel)
source("issuecomp-functions.R") source("issuecomp-functions.R")
setwd("~/Dokumente/Uni/Aktuell/BA-Arbeit/uni-ba-issuecomp")
load(file = "tweets_untagged.RData") load(file = "tweets_untagged.RData")
@@ -17,13 +18,8 @@ date_end <- as.Date("2014-12-31")
drange <- as.integer(date_end - date_start) drange <- as.integer(date_end - date_start)
drange <- date_start + days(0:drange) drange <- date_start + days(0:drange)
# Import issues and prepare everything
# MATCH TWEETS ------------------------------------------------------------ # Will only be filled after the large categorisation loop
id_folder <- "matched-ids"
#unlink(id_folder, recursive = TRUE)
#dir.create(id_folder)
issues <- data.frame(date = drange) issues <- data.frame(date = drange)
issuelist <- readLines("issues.xml") issuelist <- readLines("issues.xml")
issuelist <- str_replace_all(string = issuelist, pattern = ".*<!-- .+ -->", "") issuelist <- str_replace_all(string = issuelist, pattern = ".*<!-- .+ -->", "")
@@ -33,15 +29,24 @@ issues[issueheads] <- 0
tweets$issue <- "" tweets$issue <- ""
tweets$tags <- "" tweets$tags <- ""
# MATCH TWEETS ------------------------------------------------------------
# Create folder where all results will be saved (saver for backup and import)
id_folder <- "matched-ids"
unlink(id_folder, recursive = TRUE)
dir.create(id_folder)
# Tag expansion for plural, genetiv etc
tagexpand <- c("", "s", "n", "en", "er", "e") tagexpand <- c("", "s", "n", "en", "er", "e")
# Parallelisation # Parameters for parallelisation
writeLines(c(""), "issuecomp-analysis.log") writeLines(c(""), "issuecomp-analysis.log")
cl<-makeCluster(4) cl<-makeCluster(4)
registerDoParallel(cl) registerDoParallel(cl)
foreach(d = 260:nrow(issues), .packages = c("stringr"), .combine=rbind) %dopar% { # START CAT LOOP
#for(d in 1:nrow(issues)) { foreach(d = 1:nrow(issues), .packages = c("stringr"), .combine=rbind) %dopar% {
# Go through every day # Go through every day
curdate <- issues$date[d] curdate <- issues$date[d]
cat(paste(as.character(curdate),"\n"), file="issuecomp-analysis.log", append=TRUE) cat(paste(as.character(curdate),"\n"), file="issuecomp-analysis.log", append=TRUE)
@@ -125,12 +130,58 @@ foreach(d = 260:nrow(issues), .packages = c("stringr"), .combine=rbind) %dopar%
#rm(tweets_curday,curacro, curchars, curdate,curfile,curid,curissue,curtag,curtags,curtext,d,date_end,date_start,i,id_folder,oldissue,oldtag,s,t,tags_found) #rm(tweets_curday,curacro, curchars, curdate,curfile,curid,curissue,curtag,curtags,curtext,d,date_end,date_start,i,id_folder,oldissue,oldtag,s,t,tags_found)
stopCluster(cl) stopCluster(cl)
# IMPORT RESULTS ----------------------------------------------------------
# Import all files which have been generated at the categorisation run above.
setwd("matched-ids/")
results_files <- list.files()
for(r in 1:length(results_files)) {
if(r == 1) {
results <- read.csv(results_files[r], sep=";", colClasses=c("character", "character", "character", "character"), header=F)
} else {
results_temp <- read.csv(results_files[r], sep=";", colClasses=c("character", "character", "character", "character"), header=F)
results <- insertRow(results, results_temp)
}
}
rm(r, results_temp, results_files)
# Remove duplicates, sort chronologically
results <- results[!duplicated(results), ]
names(results) <- c("date", "id_str", "issue", "tags")
results <- results[order(results$id_str), ]
row.names(results) <- NULL
# Now import all results in the dataframes "issues" and "tweets"
# (which wasn't possible in the categorisation process because of parallelisation)
# Reset issues counter
# issues[issueheads] <- 0
for(r in 1:nrow(results)) {
curdate <- as.character(results$date[r])
curid <- as.character(results$id_str[r])
curissue <- as.character(results$issue[r])
curtag <- as.character(results$tags[r])
cat("Sorting match", r, "of 62827 \n")
# Update issue counter (date and issue)
issues[issues[, "date"] == curdate, curissue] <- issues[issues[, "date"] == curdate, curissue] + 1
# Update tweet dataframe (id, issue and tags)
oldissue <- tweets[tweets[, "id_str"] == curid, "issue"]
tweets[tweets[, "id_str"] == curid, "issue"] <- str_c(oldissue, curissue, ",")
oldtag <- tweets[tweets[, "id_str"] == curid, "tags"]
tweets[tweets[, "id_str"] == curid, "tags"] <- str_c(oldtag, curtag, ",")
}
# SAVING ------------------------------------------------------------------ # SAVING ------------------------------------------------------------------
save(tweets, file="tweets_tagged.RData")
row.names(tweets) <- NULL
write.csv(tweets, "tweets.csv")
save(tweets, file="tweets.RData")
tweets_tagged.RData
BIN
View File
Binary file not shown.