513 lines
21 KiB
R
513 lines
21 KiB
R
# 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
|
|
}
|
|
# Match current tweet with tag. If >= 5 letters allow 1 changed letter, if >=8 letters allow 2 (Levenshtein distance)
|
|
tags_found <- smartPatternMatch(curtext, curtag, curchars, curacro)
|
|
if(tags_found == 1) {
|
|
#cat("Matched", curtag, "with", curtext,"\n")
|
|
issues[d,curissue] <- issues[d,curissue] + 1
|
|
write(str_c(curdate,";\"",curid,"\""), curfile, append = TRUE)
|
|
break
|
|
}
|
|
else {
|
|
#cat("Nothing found\n")
|
|
}
|
|
} # /for curtags
|
|
} # /for issuelist
|
|
} # /for tweets_curday
|
|
} # /for drange
|
|
smartPatternMatch("kerTips: Riker workplace tip: Flirt when no one else is looking. http", "IS", 2, TRUE)
|
|
smartPatternMatch("kerTips: Riker workplace tip: Flirt when no one else is looking. http", "is", 2, TRUE)
|
|
viewMatchingTweets("2014-01-06", "issue.iraq", id_folder)
|
|
# MATCH TWEETS ------------------------------------------------------------
|
|
id_folder <- "matched-ids"
|
|
unlink(id_folder, recursive = TRUE)
|
|
dir.create(id_folder)
|
|
issues <- data.frame(date = drange)
|
|
issuelist <- xmlToList("issues.xml")
|
|
issueheads <- names(issuelist)
|
|
issues[issueheads] <- 0
|
|
for(d in 1:nrow(issues)) {
|
|
# Go through every day
|
|
curdate <- issues$date[d]
|
|
cat(as.character(curdate),"\n")
|
|
# Put all tweets from specific day in a temporary DF
|
|
tweets_curday <- tweets[tweets[, "created_at"] == curdate, ]
|
|
for(t in 1:nrow(tweets_curday)){
|
|
# Select tweet's text, make it lowercase and remove hashtag indicators (#)
|
|
curtext <- as.character(tweets_curday$text[t])
|
|
curtext <- str_replace_all(curtext, "#", "")
|
|
curid <- as.character(tweets_curday$id_str[t])
|
|
# Now test each single issue (not tag!)
|
|
for(i in 1:length(issuelist)) {
|
|
curtags <- as.character(issuelist[[i]])
|
|
curissue <- names(issuelist)[i]
|
|
curfile <- str_c(id_folder,"/",curissue,".csv")
|
|
# Now test all tags of a single issue
|
|
for(s in 1:length(curtags)) {
|
|
curtag <- curtags[s]
|
|
curchars <- nchar(curtag, type = "chars")
|
|
# Check if tag is an acronym. If so, ignore.case will be deactivated in smartPatternMatch
|
|
if(curchars <= 4) {
|
|
curacro <- checkAcronym(string = curtag, chars = curchars)
|
|
} else {
|
|
curacro <- FALSE
|
|
}
|
|
# Match current tweet with tag. If >= 5 letters allow 1 changed letter, if >=8 letters allow 2 (Levenshtein distance)
|
|
tags_found <- smartPatternMatch(curtext, curtag, curchars, curacro)
|
|
if(tags_found == 1) {
|
|
#cat("Matched", curtag, "with", curtext,"\n")
|
|
issues[d,curissue] <- issues[d,curissue] + 1
|
|
write(str_c(curdate,";\"",curid,"\""), curfile, append = TRUE)
|
|
break
|
|
}
|
|
else {
|
|
#cat("Nothing found\n")
|
|
}
|
|
} # /for curtags
|
|
} # /for issuelist
|
|
} # /for tweets_curday
|
|
} # /for drange
|
|
source("issuecomp-functions.R")
|
|
# MATCH TWEETS ------------------------------------------------------------
|
|
id_folder <- "matched-ids"
|
|
unlink(id_folder, recursive = TRUE)
|
|
dir.create(id_folder)
|
|
issues <- data.frame(date = drange)
|
|
issuelist <- xmlToList("issues.xml")
|
|
issueheads <- names(issuelist)
|
|
issues[issueheads] <- 0
|
|
for(d in 1:nrow(issues)) {
|
|
# Go through every day
|
|
curdate <- issues$date[d]
|
|
cat(as.character(curdate),"\n")
|
|
# Put all tweets from specific day in a temporary DF
|
|
tweets_curday <- tweets[tweets[, "created_at"] == curdate, ]
|
|
for(t in 1:nrow(tweets_curday)){
|
|
# 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(issuelist)) {
|
|
curtags <- as.character(issuelist[[i]])
|
|
curissue <- names(issuelist)[i]
|
|
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
|
|
}
|
|
# Match current tweet with tag. If >= 5 letters allow 1 changed letter, if >=8 letters allow 2 (Levenshtein distance)
|
|
tags_found <- smartPatternMatch(curtext, curtag, curchars, curacro)
|
|
if(tags_found == 1) {
|
|
#cat("Matched", curtag, "with", curtext,"\n")
|
|
issues[d,curissue] <- issues[d,curissue] + 1
|
|
write(str_c(curdate,";\"",curid,"\";"curtag), curfile, append = TRUE)
|
|
break
|
|
}
|
|
else {
|
|
#cat("Nothing found\n")
|
|
}
|
|
} # /for curtags
|
|
} # /for issuelist
|
|
} # /for tweets_curday
|
|
} # /for drange
|
|
# MATCH TWEETS ------------------------------------------------------------
|
|
id_folder <- "matched-ids"
|
|
unlink(id_folder, recursive = TRUE)
|
|
dir.create(id_folder)
|
|
issues <- data.frame(date = drange)
|
|
issuelist <- xmlToList("issues.xml")
|
|
issueheads <- names(issuelist)
|
|
issues[issueheads] <- 0
|
|
for(d in 1:nrow(issues)) {
|
|
# Go through every day
|
|
curdate <- issues$date[d]
|
|
cat(as.character(curdate),"\n")
|
|
# Put all tweets from specific day in a temporary DF
|
|
tweets_curday <- tweets[tweets[, "created_at"] == curdate, ]
|
|
for(t in 1:nrow(tweets_curday)){
|
|
# Select tweet's text, make it lowercase and remove hashtag indicators (#)
|
|
curtext <- as.character(tweets_curday$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(issuelist)) {
|
|
curtags <- as.character(issuelist[[i]])
|
|
curissue <- names(issuelist)[i]
|
|
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
|
|
}
|
|
# Match current tweet with tag. If >= 5 letters allow 1 changed letter, if >=8 letters allow 2 (Levenshtein distance)
|
|
tags_found <- smartPatternMatch(curtext, curtag, curchars, curacro)
|
|
if(tags_found == 1) {
|
|
#cat("Matched", curtag, "with", curtext,"\n")
|
|
issues[d,curissue] <- issues[d,curissue] + 1
|
|
write(str_c(curdate,";\"",curid,"\";",curtag), curfile, append = TRUE)
|
|
break
|
|
}
|
|
else {
|
|
#cat("Nothing found\n")
|
|
}
|
|
} # /for curtags
|
|
} # /for issuelist
|
|
} # /for tweets_curday
|
|
} # /for drange
|
|
source("issuecomp-functions.R")
|
|
viewMatchingTweets("2014-01-06", "issue.iraq", id_folder)
|
|
viewMatchingTweets("2014-01-07", "issue.iraq", id_folder)
|
|
viewMatchingTweets("2014-01-09", "issue.iraq", id_folder)
|
|
curtext <- "Willkürlich Menschen an ihrer #Versammlungsfreiheit zu hindern ist eindeutig rechtswidrig. http://t.co/A7IQfISIhP #Gefahrengebiet #Hamburg"
|
|
str_replace_all(curtext, "http://.+\\W", "")
|
|
str_replace_all(curtext, "http://.+?\\W", "")
|
|
str_replace_all(curtext, "http://.+?\\s", "")
|
|
str_replace_all(curtext, "http://.+?\\s", "")
|
|
curtext <- "test http://google.de haha http://nsa.gov eqiuhe"
|
|
str_replace_all(curtext, "http://.+?\\s", "")
|
|
str_replace_all(curtext, "http://.+?\\s", "URL")
|
|
str_replace_all(curtext, "http://.+?\\s", "URL ")
|
|
viewMatchingTweets("2014-01-09", "issue.iraq", id_folder)
|
|
# MATCH TWEETS ------------------------------------------------------------
|
|
id_folder <- "matched-ids"
|
|
unlink(id_folder, recursive = TRUE)
|
|
dir.create(id_folder)
|
|
issues <- data.frame(date = drange)
|
|
issuelist <- xmlToList("issues.xml")
|
|
issueheads <- names(issuelist)
|
|
issues[issueheads] <- 0
|
|
for(d in 1:nrow(issues)) {
|
|
# Go through every day
|
|
curdate <- issues$date[d]
|
|
cat(as.character(curdate),"\n")
|
|
# Put all tweets from specific day in a temporary DF
|
|
tweets_curday <- tweets[tweets[, "created_at"] == curdate, ]
|
|
for(t in 1:nrow(tweets_curday)){
|
|
# Select tweet's text, make it lowercase and remove hashtag indicators (#)
|
|
curtext <- as.character(tweets_curday$text[t])
|
|
curtext <- str_replace_all(curtext, "#", "")
|
|
curtext <- str_replace_all(curtext, "http://.+?\\s", "URL ")
|
|
curid <- as.character(tweets_curday$id_str[t])
|
|
# Now test each single issue (not tag!)
|
|
for(i in 1:length(issuelist)) {
|
|
curtags <- as.character(issuelist[[i]])
|
|
curissue <- names(issuelist)[i]
|
|
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
|
|
}
|
|
# Match current tweet with tag. If >= 5 letters allow 1 changed letter, if >=8 letters allow 2 (Levenshtein distance)
|
|
tags_found <- smartPatternMatch(curtext, curtag, curchars, curacro)
|
|
if(tags_found == 1) {
|
|
#cat("Matched", curtag, "with", curtext,"\n")
|
|
issues[d,curissue] <- issues[d,curissue] + 1
|
|
write(str_c(curdate,";\"",curid,"\";",curtag), curfile, append = TRUE)
|
|
break
|
|
}
|
|
else {
|
|
#cat("Nothing found\n")
|
|
}
|
|
} # /for curtags
|
|
} # /for issuelist
|
|
} # /for tweets_curday
|
|
} # /for drange
|
|
viewMatchingTweets("2014-01-09", "issue.iraq", id_folder)
|
|
viewMatchingTweets("2014-01-08", "issue.iraq", id_folder)
|
|
viewMatchingTweets("2014-01-10", "issue.iraq", id_folder)
|
|
curtext
|
|
str_replace_all(curtext, "http://.+?\\>", "URL ")
|
|
str_replace_all(curtext, "http://.+?\\<", "URL ")
|
|
curtext <- str_replace_all(curtext, "http://.+?\\b", "URL ")
|
|
str_replace_all(curtext, "http://.+?\\b", "URL ")
|
|
str_replace_all(curtext, "http://.+?\\s", "URL ")
|
|
curtext
|
|
curtext <- as.character(tweets_curday$text[t])
|
|
curtext
|
|
str_replace_all(curtext, "http://.+?\\s", "URL ")
|
|
str_replace_all(curtext, "http://.+?\\b", "URL ")
|
|
str_replace_all(curtext, "http://.+?\\<", "URL ")
|
|
str_replace_all(curtext, "http://.+?\\>", "URL ")
|
|
str_replace_all(curtext, "http://.+?\\s", "URL ")
|
|
str_replace_all(curtext, "$", " ")
|
|
curtext <- str_replace_all(curtext, "$", " ")
|
|
curtext
|
|
str_replace_all(curtext, "http://.+?\\s", "URL ")
|
|
viewMatchingTweets("2014-01-10", "issue.iraq", id_folder)
|
|
# MATCH TWEETS ------------------------------------------------------------
|
|
id_folder <- "matched-ids"
|
|
unlink(id_folder, recursive = TRUE)
|
|
dir.create(id_folder)
|
|
issues <- data.frame(date = drange)
|
|
issuelist <- xmlToList("issues.xml")
|
|
issueheads <- names(issuelist)
|
|
issues[issueheads] <- 0
|
|
for(d in 1:nrow(issues)) {
|
|
# Go through every day
|
|
curdate <- issues$date[d]
|
|
cat(as.character(curdate),"\n")
|
|
# Put all tweets from specific day in a temporary DF
|
|
tweets_curday <- tweets[tweets[, "created_at"] == curdate, ]
|
|
for(t in 1:nrow(tweets_curday)){
|
|
# Select tweet's text, make it lowercase and remove hashtag indicators (#)
|
|
curtext <- as.character(tweets_curday$text[t])
|
|
curtext <- str_replace_all(curtext, "#", "")
|
|
curtext <- str_replace_all(curtext, "$", " ")
|
|
curtext <- str_replace_all(curtext, "http://.+?\\s", "URL ")
|
|
curid <- as.character(tweets_curday$id_str[t])
|
|
# Now test each single issue (not tag!)
|
|
for(i in 1:length(issuelist)) {
|
|
curtags <- as.character(issuelist[[i]])
|
|
curissue <- names(issuelist)[i]
|
|
curfile <- str_c(id_folder,"/",curissue,".csv")
|
|
# Now test all tags of a single issue
|
|
for(s in 1:length(curtags)) {
|
|
curtag <- curtags[s]
|
|
curchars <- nchar(curtag, type = "chars")
|
|
# Check if tag is an acronym. If so, ignore.case will be deactivated in smartPatternMatch
|
|
if(curchars <= 4) {
|
|
curacro <- checkAcronym(string = curtag, chars = curchars)
|
|
} else {
|
|
curacro <- FALSE
|
|
}
|
|
# Match current tweet with tag. If >= 5 letters allow 1 changed letter, if >=8 letters allow 2 (Levenshtein distance)
|
|
tags_found <- smartPatternMatch(curtext, curtag, curchars, curacro)
|
|
if(tags_found == 1) {
|
|
#cat("Matched", curtag, "with", curtext,"\n")
|
|
issues[d,curissue] <- issues[d,curissue] + 1
|
|
write(str_c(curdate,";\"",curid,"\";",curtag), curfile, append = TRUE)
|
|
break
|
|
}
|
|
else {
|
|
#cat("Nothing found\n")
|
|
}
|
|
} # /for curtags
|
|
} # /for issuelist
|
|
} # /for tweets_curday
|
|
} # /for drange
|
|
viewMatchingTweets("2014-01-10", "issue.iraq", id_folder)
|
|
# MATCH TWEETS ------------------------------------------------------------
|
|
id_folder <- "matched-ids"
|
|
unlink(id_folder, recursive = TRUE)
|
|
dir.create(id_folder)
|
|
issues <- data.frame(date = drange)
|
|
issuelist <- xmlToList("issues.xml")
|
|
issueheads <- names(issuelist)
|
|
issues[issueheads] <- 0
|
|
for(d in 1:nrow(issues)) {
|
|
# Go through every day
|
|
curdate <- issues$date[d]
|
|
cat(as.character(curdate),"\n")
|
|
# Put all tweets from specific day in a temporary DF
|
|
tweets_curday <- tweets[tweets[, "created_at"] == curdate, ]
|
|
for(t in 1:nrow(tweets_curday)){
|
|
# Select tweet's text, make it lowercase and remove hashtag indicators (#)
|
|
curtext <- as.character(tweets_curday$text[t])
|
|
curtext <- str_replace_all(curtext, "#", "")
|
|
curtext <- str_replace_all(curtext, "$", " ")
|
|
curtext <- str_replace_all(curtext, "http://.+?\\s", "URL ")
|
|
curid <- as.character(tweets_curday$id_str[t])
|
|
# Now test each single issue (not tag!)
|
|
for(i in 1:length(issuelist)) {
|
|
curtags <- as.character(issuelist[[i]])
|
|
curissue <- names(issuelist)[i]
|
|
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
|
|
}
|
|
# Match current tweet with tag. If >= 5 letters allow 1 changed letter, if >=8 letters allow 2 (Levenshtein distance)
|
|
tags_found <- smartPatternMatch(curtext, curtag, curchars, curacro)
|
|
if(tags_found == 1) {
|
|
#cat("Matched", curtag, "with", curtext,"\n")
|
|
issues[d,curissue] <- issues[d,curissue] + 1
|
|
write(str_c(curdate,";\"",curid,"\";",curtag), curfile, append = TRUE)
|
|
break
|
|
}
|
|
else {
|
|
#cat("Nothing found\n")
|
|
}
|
|
} # /for curtags
|
|
} # /for issuelist
|
|
} # /for tweets_curday
|
|
} # /for drange
|
|
View(issues)
|
|
viewMatchingTweets("2014-12-18", "issue.edathy", id_folder)
|
|
issues_melt <- melt(issues,id="date")
|
|
ggplot(issues_melt,aes(x=date,y=value,colour=variable,group=variable)) + geom_line(size=1)
|
|
ggplot(issues_melt,aes(x=date,y=value,colour=variable,group=variable)) + geom_smooth(size=1,method="loess",formula = y ~ x, se=FALSE)
|
|
ggplot(issues_melt,aes(x=date,y=value,colour=variable,group=variable)) + geom_line(size=1)
|
|
ggplot(issues_melt,aes(x=date,y=value,colour=variable,group=variable)) + geom_line(size=1)
|
|
ggplot(issues_melt,aes(x=date,y=value,colour=variable,group=variable)) + geom_smooth(size=1,method="loess",formula = y ~ x, se=FALSE)
|
|
viewMatchingTweets("2014-12-18", "issue.conservative", id_folder)
|
|
agrep("christ", "Jungparlamentarier gleich Schriftführerdienst hat", max.distance = list(all = 2), ignore.case = TRUE, fixed = FALSE)
|
|
agrep("\\bchrist\\b", "Jungparlamentarier gleich Schriftführerdienst hat", max.distance = list(all = 2), ignore.case = TRUE, fixed = FALSE)
|
|
agrep("\\bchrist\\b", "Bla Christ bla", max.distance = list(all = 2), ignore.case = TRUE, fixed = FALSE)
|
|
agrep("\\bchrist\\b", "Bla Christus bla", max.distance = list(all = 2), ignore.case = TRUE, fixed = FALSE)
|
|
agrep("\\bchrist\\b", "Bla Christu bla", max.distance = list(all = 2), ignore.case = TRUE, fixed = FALSE)
|
|
agrep("\\bchrist\\b", "Bla Christus bla", max.distance = list(all = 2), ignore.case = TRUE, fixed = FALSE)
|
|
agrep("\\bchrist\\b", "Bla Christus bla", max.distance = list(all = 3), ignore.case = TRUE, fixed = FALSE)
|
|
agrep("\\bchrist\\b", "Bla christus bla", max.distance = list(all = 3), ignore.case = TRUE, fixed = FALSE)
|
|
agrep("\\bchrist\\b", "Bla christus bla", max.distance = list(all = 2), ignore.case = TRUE, fixed = FALSE)
|
|
agrep("\\bchrist\\b", "Bla christen bla", max.distance = list(all = 3), ignore.case = TRUE, fixed = FALSE)
|
|
agrep("\\bchrist\\b", "Bla Antichrist bla", max.distance = list(all = 3), ignore.case = TRUE, fixed = FALSE)
|
|
agrep("\\bchrist\\b", "Bla Christian bla", max.distance = list(all = 3), ignore.case = TRUE, fixed = FALSE)
|
|
agrep("\\bchrist\\b", "Bla Christian bla", max.distance = list(all = 3), ignore.case = TRUE, fixed = FALSE, value=TRUE)
|
|
agrep("\\bchrist\\b", "Bla Christi bla", max.distance = list(all = 3), ignore.case = TRUE, fixed = FALSE, value=TRUE)
|
|
agrep("\\bchrist\\b", "Bla Christi bla", max.distance = list(all = 3), ignore.case = TRUE, fixed = FALSE)
|
|
agrep("\\bIS\\b", "Wir sind bei ISN Network", max.distance = list(all = 0), ignore.case = TRUE, fixed = FALSE)
|
|
agrep("\\bIS\\b", "Wir sind bei ISN Network", max.distance = list(all = 0), ignore.case = F, fixed = FALSE)
|
|
id_folder <- "matched-ids"
|
|
unlink(id_folder, recursive = TRUE)
|
|
dir.create(id_folder)
|
|
issues <- data.frame(date = drange)
|
|
issuelist <- xmlToList("issues.xml")
|
|
issueheads <- names(issuelist)
|
|
issues[issueheads] <- 0
|
|
for(d in 1:nrow(issues)) {
|
|
# Go through every day
|
|
curdate <- issues$date[d]
|
|
cat(as.character(curdate),"\n")
|
|
# Put all tweets from specific day in a temporary DF
|
|
tweets_curday <- tweets[tweets[, "created_at"] == curdate, ]
|
|
for(t in 1:nrow(tweets_curday)){
|
|
# Select tweet's text, make it lowercase and remove hashtag indicators (#)
|
|
curtext <- as.character(tweets_curday$text[t])
|
|
curtext <- str_replace_all(curtext, "#", "")
|
|
curtext <- str_replace_all(curtext, "$", " ")
|
|
curtext <- str_replace_all(curtext, "http://.+?\\s", "URL ")
|
|
curid <- as.character(tweets_curday$id_str[t])
|
|
# Now test each single issue (not tag!)
|
|
for(i in 1:length(issuelist)) {
|
|
curtags <- as.character(issuelist[[i]])
|
|
curissue <- names(issuelist)[i]
|
|
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
|
|
}
|
|
# Match current tweet with tag. If >= 5 letters allow 1 changed letter, if >=8 letters allow 2 (Levenshtein distance)
|
|
tags_found <- smartPatternMatch(curtext, curtag, curchars, curacro)
|
|
if(tags_found == 1) {
|
|
#cat("Matched", curtag, "with", curtext,"\n")
|
|
issues[d,curissue] <- issues[d,curissue] + 1
|
|
write(str_c(curdate,";\"",curid,"\";",curtag), curfile, append = TRUE)
|
|
break
|
|
}
|
|
else {
|
|
#cat("Nothing found\n")
|
|
}
|
|
} # /for curtags
|
|
} # /for issuelist
|
|
} # /for tweets_curday
|
|
} # /for drange
|
|
issues_melt <- melt(issues,id="date")
|
|
ggplot(issues_melt,aes(x=date,y=value,colour=variable,group=variable)) + geom_smooth(size=1,method="loess",formula = y ~ x, se=FALSE)
|
|
viewMatchingTweets("2014-12-18", "issue.conservative", id_folder)
|
|
pattern
|
|
agrep("\\bchrist\\b", "RT @christophheyes: Morgen in der Presse: Oppermann - Briefkasten gestohlen! Gabriel - Poesiealbum nicht mehr auffindbar! #edathy #hartmann", max.distance = list(all = 1), ignore.case = TRUE, fixed = FALSE)
|
|
smartPatternMatch
|
|
source("issuecomp-functions.R")
|
|
smartPatternMatch
|
|
# MATCH TWEETS ------------------------------------------------------------
|
|
id_folder <- "matched-ids"
|
|
unlink(id_folder, recursive = TRUE)
|
|
dir.create(id_folder)
|
|
issues <- data.frame(date = drange)
|
|
issuelist <- xmlToList("issues.xml")
|
|
issueheads <- names(issuelist)
|
|
issues[issueheads] <- 0
|
|
for(d in 1:nrow(issues)) {
|
|
# Go through every day
|
|
curdate <- issues$date[d]
|
|
cat(as.character(curdate),"\n")
|
|
# Put all tweets from specific day in a temporary DF
|
|
tweets_curday <- tweets[tweets[, "created_at"] == curdate, ]
|
|
for(t in 1:nrow(tweets_curday)){
|
|
# Select tweet's text, make it lowercase and remove hashtag indicators (#)
|
|
curtext <- as.character(tweets_curday$text[t])
|
|
curtext <- str_replace_all(curtext, "#", "")
|
|
curtext <- str_replace_all(curtext, "$", " ")
|
|
curtext <- str_replace_all(curtext, "http://.+?\\s", "URL ")
|
|
curid <- as.character(tweets_curday$id_str[t])
|
|
# Now test each single issue (not tag!)
|
|
for(i in 1:length(issuelist)) {
|
|
curtags <- as.character(issuelist[[i]])
|
|
curissue <- names(issuelist)[i]
|
|
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
|
|
}
|
|
# Match current tweet with tag. If >= 5 letters allow 1 changed letter, if >=8 letters allow 2 (Levenshtein distance)
|
|
tags_found <- smartPatternMatch(curtext, curtag, curchars, curacro)
|
|
if(tags_found == 1) {
|
|
#cat("Matched", curtag, "with", curtext,"\n")
|
|
issues[d,curissue] <- issues[d,curissue] + 1
|
|
write(str_c(curdate,";\"",curid,"\";",curtag), curfile, append = TRUE)
|
|
break
|
|
}
|
|
else {
|
|
#cat("Nothing found\n")
|
|
}
|
|
} # /for curtags
|
|
} # /for issuelist
|
|
} # /for tweets_curday
|
|
} # /for drange
|
|
issues_melt <- melt(issues,id="date")
|
|
ggplot(issues_melt,aes(x=date,y=value,colour=variable,group=variable)) + geom_smooth(size=1,method="loess",formula = y ~ x, se=FALSE)
|
|
viewMatchingTweets("2014-12-18", "issue.conservative", id_folder)
|
|
viewMatchingTweets("2014-05-18", "issue.conservative", id_folder)
|
|
viewMatchingTweets("2014-05-1", "issue.conservative", id_folder)
|
|
viewMatchingTweets("2014-05-01", "issue.conservative", id_folder)
|
|
viewMatchingTweets("2014-05-02", "issue.conservative", id_folder)
|
|
viewMatchingTweets("2014-05-10", "issue.conservative", id_folder)
|
|
viewMatchingTweets("2014-05-10", "issue.middleeast", id_folder)
|
|
viewMatchingTweets("2014-05-10", "issue.iraw", id_folder)
|
|
viewMatchingTweets("2014-05-10", "issue.iraq", id_folder)
|
|
viewMatchingTweets("2014-08-10", "issue.iraq", id_folder)
|
|
viewMatchingTweets("2014-11-10", "issue.iraq", id_folder)
|
|
viewMatchingTweets("2014-12-10", "issue.iraq", id_folder)
|
|
View(issues)
|
|
viewMatchingTweets("2014-09-19", "issue.control", id_folder)
|