From b1a6a548f04987b367408b2c328296a3a4609695 Mon Sep 17 00:00:00 2001
From: mxmehl <mail@mehl.mx>
Date: Sun, 22 Feb 2015 03:18:51 +0100
Subject: [PATCH] output still has to be directed to correct data frame

---
 .Rhistory              | 924 ++++++++++++++++++++---------------------
 issuecomp-analysis.R   |   8 +-
 issuecomp-analysis.log | 435 +++----------------
 3 files changed, 518 insertions(+), 849 deletions(-)

diff --git a/.Rhistory b/.Rhistory
index 48415fe..d55074d 100644
--- a/.Rhistory
+++ b/.Rhistory
@@ -1,465 +1,3 @@
-#loop
-ls<-foreach(i = 1:length(drange)) %dopar% {
-sink("log.txt", append=TRUE)
-as.character(drange[i])
-w <- sample(1:2, 1)
-Sys.sleep(w)
-}
-stopCluster(cl)
-#import packages
-library(foreach)
-library(doParallel)
-#setup parallel backend to use 8 processors
-cl<-makeCluster(3)
-registerDoParallel(cl)
-#start time
-strt<-Sys.time()
-writeLines(c(""), "log.txt")
-#loop
-ls<-foreach(i = 1:length(drange)) %dopar% {
-sink("log.txt", append=TRUE)
-as.character(drange[i])
-}
-print(Sys.time()-strt)
-stopCluster(cl)
-#import packages
-library(foreach)
-library(doParallel)
-#setup parallel backend to use 8 processors
-cl<-makeCluster(3)
-registerDoParallel(cl)
-#start time
-strt<-Sys.time()
-writeLines(c(""), "log.txt")
-#loop
-ls<-foreach(i = 1:length(drange)) %dopar% {
-sink("log.txt", append=TRUE)
-cat(as.character(drange[i]))
-}
-print(Sys.time()-strt)
-stopCluster(cl)
-writeLines(c(""), "log.txt")
-cat(as.character(drange[i]))
-writeLines(c(""), "log.txt")
-#import packages
-library(foreach)
-library(doParallel)
-#setup parallel backend to use 8 processors
-cl<-makeCluster(3)
-registerDoParallel(cl)
-#start time
-strt<-Sys.time()
-writeLines(c(""), "log.txt")
-#loop
-ls<-foreach(i = 1:length(drange)) %dopar% {
-sink("log.txt", append=TRUE)
-cat(as.character(drange[i]),"\n")
-}
-print(Sys.time()-strt)
-stopCluster(cl)
-#import packages
-library(foreach)
-library(doParallel)
-#setup parallel backend to use 8 processors
-cl<-makeCluster(3)
-registerDoParallel(cl)
-#start time
-strt<-Sys.time()
-writeLines(c(""), "log.txt")
-#loop
-ls<-foreach(i = 1:length(drange)) %dopar% {
-sink("log.txt", append=TRUE)
-cat(as.character(drange[i]),"\n")
-w <- sample(1:3, 1)
-Sys.sleep(w)
-as.character(drange[i])
-}
-print(Sys.time()-strt)
-stopCluster(cl)
-#import packages
-library(foreach)
-library(doParallel)
-#setup parallel backend to use 8 processors
-cl<-makeCluster(3)
-registerDoParallel(cl)
-#start time
-strt<-Sys.time()
-writeLines(c(""), "log.txt")
-#loop
-ls<-foreach(i = 1:length(drange)) %dopar% {
-sink("log.txt", append=TRUE)
-cat(as.character(drange[i]),"\n")
-#   w <- sample(1:3, 1)
-#   Sys.sleep(w)
-as.character(drange[i])
-}
-print(Sys.time()-strt)
-stopCluster(cl)
-#import packages
-library(foreach)
-library(doParallel)
-#setup parallel backend to use 8 processors
-cl<-makeCluster(3)
-registerDoParallel(cl)
-#start time
-strt<-Sys.time()
-writeLines(c(""), "log.txt")
-#loop
-ls<-foreach(i = 1:length(drange)) %dopar% {
-sink("log.txt", append=TRUE)
-cat(as.character(drange[i]),"\n")
-#   w <- sample(1:3, 1)
-#   Sys.sleep(w)
-as.character(drange[i])
-}
-print(Sys.time()-strt)
-stopCluster(cl)
-#import packages
-library(foreach)
-library(doParallel)
-#setup parallel backend to use 8 processors
-cl<-makeCluster(3)
-registerDoParallel(cl)
-#start time
-strt<-Sys.time()
-writeLines(c(""), "log.txt")
-#loop
-ls<-foreach(i = 1:length(drange)) %dopar% {
-cat(paste("\n","Starting iteration",i,"\n"), file="log.txt", append=TRUE)
-as.character(drange[i])
-}
-print(Sys.time()-strt)
-stopCluster(cl)
-#import packages
-library(foreach)
-library(doParallel)
-#setup parallel backend to use 8 processors
-cl<-makeCluster(3)
-registerDoParallel(cl)
-#start time
-strt<-Sys.time()
-writeLines(c(""), "log.txt")
-#loop
-ls<-foreach(i = 1:length(drange)) %dopar% {
-w <- sample(1:3, 1)
-Sys.sleep(w)
-cat(paste("\n","Starting iteration",i,"\n"), file="log.txt", append=TRUE)
-as.character(drange[i])
-}
-#import packages
-library(foreach)
-library(doParallel)
-#setup parallel backend to use 8 processors
-cl<-makeCluster(3)
-registerDoParallel(cl)
-#start time
-strt<-Sys.time()
-writeLines(c(""), "log.txt")
-#loop
-ls<-foreach(i = 1:length(drange)) %dopar% {
-w <- sample(1:10, 1)
-Sys.sleep(w)
-cat(paste("\n","Starting iteration",i,"\n"), file="log.txt", append=TRUE)
-as.character(drange[i])
-}
-#import packages
-library(foreach)
-library(doParallel)
-#setup parallel backend to use 8 processors
-cl<-makeCluster(3)
-registerDoParallel(cl)
-#start time
-strt<-Sys.time()
-writeLines(c(""), "log.txt")
-#loop
-ls<-foreach(i = 1:length(drange)) %dopar% {
-w <- sample(1:10, 1)
-#Sys.sleep(w)
-cat(paste("\n","Starting iteration",i,"\n"), file="log.txt", append=TRUE)
-as.character(drange[i])
-}
-print(Sys.time()-strt)
-stopCluster(cl)
-#import packages
-library(foreach)
-library(doParallel)
-#setup parallel backend to use 8 processors
-cl<-makeCluster(3)
-registerDoParallel(cl)
-#start time
-strt<-Sys.time()
-writeLines(c(""), "log.txt")
-#loop
-ls<-foreach(i = 1:length(drange)) %dopar% {
-w <- sample(1:10, 1)
-#Sys.sleep(w)
-cat(paste("\n","Starting iteration",i,"\n"), file="log.txt", append=TRUE)
-as.character(drange[i])
-}
-print(Sys.time()-strt)
-stopCluster(cl)
-View(data)
-#import packages
-library(foreach)
-library(doParallel)
-#setup parallel backend to use 8 processors
-cl<-makeCluster(3)
-registerDoParallel(cl)
-#start time
-strt<-Sys.time()
-writeLines(c(""), "log.txt")
-#loop
-data<-foreach(i = 1:length(drange)) %dopar% {
-w <- sample(1:10, 1)
-#Sys.sleep(w)
-cat(paste("\n","Starting iteration",i,"\n"), file="log.txt", append=TRUE)
-as.character(drange[i])
-}
-print(Sys.time()-strt)
-stopCluster(cl)
-rm(ls)
-data
-#import packages
-library(foreach)
-library(doParallel)
-#setup parallel backend to use 8 processors
-cl<-makeCluster(3)
-registerDoParallel(cl)
-#start time
-strt<-Sys.time()
-writeLines(c(""), "log.txt")
-#loop
-df<-foreach(i = 1:length(drange)) %dopar% {
-w <- sample(1:10, 1)
-#Sys.sleep(w)
-cat(paste("\n","Starting iteration",i,"\n"), file="log.txt", append=TRUE)
-as.character(drange[i])
-}
-print(Sys.time()-strt)
-stopCluster(cl)
-df
-view(df)
-View(df)
-# Parallelisation
-writeLines(c(""), "log.txt")
-cl<-makeCluster(3)
-registerDoParallel(cl)
-# 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")
-# Parallelisation
-writeLines(c(""), "issuecomp-analysis.log")
-cl<-makeCluster(3)
-registerDoParallel(cl)
-df<-foreach(d = 1:nrow(issues) %dopar% {
-#for(d in 1:nrow(issues)) {
-# Go through every day
-curdate <- issues$date[d]
-sink("log.txt", append=TRUE)
-cat(as.character(curdate),"\n")
-# Put all tweets from specific day in a temporary DF
-tweets_curday <- tweets[tweets[, "created_at"] == curdate, ]
-for(t in 1:nrow(tweets_curday)){
-cat("Starting tweet", t, "of",as.character(curdate),"\n")
-# Select tweet's text, make it lowercase and remove hashtag indicators (#)
-curtext <- as.character(tweets_curday$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,"\";",curtag), curfile, append = TRUE)
-cat("Match!\n")
-break
-}
-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)
-# 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$tags <- ""
-tagexpand <- c("", "s", "n", "en", "er")
-# Parallelisation
-writeLines(c(""), "issuecomp-analysis.log")
-cl<-makeCluster(3)
-registerDoParallel(cl)
-df<-foreach(d = 1:nrow(issues)) %dopar% {
-#for(d in 1:nrow(issues)) {
-# Go through every day
-curdate <- issues$date[d]
-sink("log.txt", append=TRUE)
-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)){
-cat("Starting tweet", t, "of",as.character(curdate),"\n")
-# 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,"\";",curtag), curfile, append = TRUE)
-cat("Match!\n")
-break
-}
-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)
-# 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")
-# Parallelisation
-writeLines(c(""), "issuecomp-analysis.log")
-cl<-makeCluster(3)
-registerDoParallel(cl)
-df<-foreach(d = 1:nrow(issues), .packages = c("stringr")) %dopar% {
-#for(d in 1:nrow(issues)) {
-# Go through every day
-curdate <- issues$date[d]
-sink("log.txt", append=TRUE)
-cat(as.character(curdate),"\n")
-# Put all tweets from specific day in a temporary DF
-tweets_curday <- tweets[tweets[, "created_at"] == curdate, ]
-for(t in 1:nrow(tweets_curday)){
-cat("Starting tweet", t, "of",as.character(curdate),"\n")
-# Select tweet's text, make it lowercase and remove hashtag indicators (#)
-curtext <- as.character(tweets_curday$text[t])
-curtext <- str_replace_all(curtext, "#", "")
-curid <- as.character(tweets_curday$id_str[t])
-# Now test each single issue (not tag!)
-for(i in 1:length(issueheads)) {
-curissue <- issueheads[i]
-curtags <- as.character(issuelist[[curissue]])
-curfile <- str_c(id_folder,"/",curissue,".csv")
-# Now test all tags of a single issue
-for(s in 1:length(curtags)) {
-curtag <- curtags[s]
-curchars <- nchar(curtag, type = "chars")
-# Check if tag is an acronym. If so, ignore.case will be deactivated in smartPatternMatch
-if(curchars <= 4) {
 curacro <- checkAcronym(string = curtag, chars = curchars)
 } else {
 curacro <- FALSE
@@ -510,3 +48,465 @@ cl
 df
 View(data)
 stopCluster(cl)
+# MATCH TWEETS ------------------------------------------------------------
+id_folder <- "matched-ids"
+unlink(id_folder, recursive = TRUE)
+dir.create(id_folder)
+issues <- data.frame(date = drange)
+issuelist <- readLines("issues.xml")
+issuelist <- str_replace_all(string = issuelist, pattern = ".*<!-- .+ -->", "")
+issuelist <- xmlToList(issuelist)
+issueheads <- names(issuelist)
+issues[issueheads] <- 0
+tweets$issue <- ""
+tweets$tags <- ""
+tagexpand <- c("", "s", "n", "en", "er")
+# Parallelisation
+writeLines(c(""), "issuecomp-analysis.log")
+cl<-makeCluster(3)
+registerDoParallel(cl)
+df<-foreach(d = 1:nrow(issues), .packages = c("stringr")) %dopar% {
+#for(d in 1:nrow(issues)) {
+# Go through every day
+curdate <- issues$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,"\";",curtag), curfile, append = TRUE)
+cat("Match!\n")
+break
+}
+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)
+require(lubridate)
+require(XML)
+require(ggplot2)
+require(reshape2)
+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")
+# Parallelisation
+writeLines(c(""), "issuecomp-analysis.log")
+cl<-makeCluster(3)
+registerDoParallel(cl)
+df<-foreach(d = 1:nrow(issues), .packages = c("stringr")) %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,"\";",curtag), curfile, append = TRUE)
+cat("Match!\n")
+break
+}
+else {
+#cat("Nothing found\n")
+}
+} # /for curtags
+} # /for issuelist
+} # /for tweets_curday
+} # /for drange
+stopCluster(cl)
+cl<-makeCluster(3)
+registerDoParallel(cl)
+stopCluster(cl)
+# 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$tags <- ""
+tagexpand <- c("", "s", "n", "en", "er")
+# Parallelisation
+writeLines(c(""), "issuecomp-analysis.log")
+cl<-makeCluster(3)
+registerDoParallel(cl)
+df<-foreach(d = 1:nrow(issues), .packages = c("stringr"), .combine=rbind) %dopar% {
+#for(d in 1:nrow(issues)) {
+# Go through every day
+curdate <- issues$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,"\";",curtag), curfile, append = TRUE)
+cat("Match!\n")
+break
+}
+else {
+#cat("Nothing found\n")
+}
+} # /for curtags
+} # /for issuelist
+} # /for tweets_curday
+} # /for drange
+stopCluster(cl)
+require(lubridate)
+require(XML)
+require(ggplot2)
+require(reshape2)
+require(stringr)
+library(foreach)
+library(doParallel)
+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 <- 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")
+# Parallelisation
+writeLines(c(""), "issuecomp-analysis.log")
+cl<-makeCluster(3)
+registerDoParallel(cl)
+df<-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,"\";",curtag), curfile, append = TRUE)
+cat(paste("Match!\n"), file="issuecomp-analysis.log", append=TRUE)
+break
+}
+else {
+#cat("Nothing found\n")
+}
+} # /for curtags
+} # /for issuelist
+} # /for tweets_curday
+} # /for drange
+View(issues)
+require(lubridate)
+require(XML)
+require(ggplot2)
+require(reshape2)
+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$tags <- ""
+tagexpand <- c("", "s", "n", "en", "er")
+# Parallelisation
+writeLines(c(""), "issuecomp-analysis.log")
+cl<-makeCluster(3)
+registerDoParallel(cl)
+df<-foreach(d = 1:3, .packages = c("stringr"), .combine=rbind) %dopar% {
+#df<-foreach(d = 1:nrow(issues), .packages = c("stringr"), .combine=rbind) %dopar% {
+#for(d in 1:nrow(issues)) {
+# Go through every day
+curdate <- issues$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 1:25){
+#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,"\";",curtag), curfile, append = TRUE)
+cat(paste("Match!\n"), file="issuecomp-analysis.log", append=TRUE)
+break
+}
+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)
+View(issues)
+rm(data)
+df
diff --git a/issuecomp-analysis.R b/issuecomp-analysis.R
index 2561858..c33f18e 100644
--- a/issuecomp-analysis.R
+++ b/issuecomp-analysis.R
@@ -40,7 +40,8 @@ writeLines(c(""), "issuecomp-analysis.log")
 cl<-makeCluster(3)
 registerDoParallel(cl)
 
-df<-foreach(d = 1:nrow(issues), .packages = c("stringr"), .combine=rbind) %dopar% {
+df<-foreach(d = 1:3, .packages = c("stringr"), .combine=rbind) %dopar% {
+#df<-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]
@@ -49,7 +50,8 @@ df<-foreach(d = 1:nrow(issues), .packages = c("stringr"), .combine=rbind) %dopar
   # Put all tweets from specific day in a temporary DF
   tweets_curday <- tweets[tweets[, "created_at"] == curdate, ]
   
-  for(t in 1:nrow(tweets_curday)){
+  for(t in 1:25){
+  #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])
@@ -110,7 +112,7 @@ df<-foreach(d = 1:nrow(issues), .packages = c("stringr"), .combine=rbind) %dopar
           
           # Add information to file for function viewPatternMatching
           write(str_c(curdate,";\"",curid,"\";",curtag), curfile, append = TRUE)
-          cat("Match!\n")
+          cat(paste("Match!\n"), file="issuecomp-analysis.log", append=TRUE)
           break
         }
         else {
diff --git a/issuecomp-analysis.log b/issuecomp-analysis.log
index 8a76877..c073cf9 100644
--- a/issuecomp-analysis.log
+++ b/issuecomp-analysis.log
@@ -2,435 +2,102 @@
 2014-01-01 
 2014-01-02 
 2014-01-03 
-Starting tweet 1 of 2014-01-02 
-Starting tweet 1 of 2014-01-01 
 Starting tweet 1 of 2014-01-03 
+Starting tweet 1 of 2014-01-01 
+Starting tweet 1 of 2014-01-02 
 Starting tweet 2 of 2014-01-01 
 Starting tweet 2 of 2014-01-03 
 Starting tweet 2 of 2014-01-02 
-Starting tweet 3 of 2014-01-01 
 Starting tweet 3 of 2014-01-03 
+Starting tweet 3 of 2014-01-01 
+Match!
 Starting tweet 3 of 2014-01-02 
-Starting tweet 4 of 2014-01-01 
 Starting tweet 4 of 2014-01-03 
+Match!
+Starting tweet 4 of 2014-01-01 
+Match!
 Starting tweet 4 of 2014-01-02 
-Starting tweet 5 of 2014-01-01 
 Starting tweet 5 of 2014-01-03 
-Starting tweet 6 of 2014-01-01 
-Starting tweet 5 of 2014-01-02 
+Starting tweet 5 of 2014-01-01 
 Starting tweet 6 of 2014-01-03 
+Match!
+Starting tweet 5 of 2014-01-02 
+Match!
+Starting tweet 6 of 2014-01-01 
 Starting tweet 6 of 2014-01-02 
-Starting tweet 7 of 2014-01-01 
 Starting tweet 7 of 2014-01-03 
-Starting tweet 8 of 2014-01-01 
+Starting tweet 7 of 2014-01-01 
+Match!
 Starting tweet 7 of 2014-01-02 
 Starting tweet 8 of 2014-01-03 
-Starting tweet 9 of 2014-01-01 
+Starting tweet 8 of 2014-01-01 
+Match!
 Starting tweet 8 of 2014-01-02 
 Starting tweet 9 of 2014-01-03 
+Starting tweet 9 of 2014-01-01 
 Starting tweet 9 of 2014-01-02 
-Starting tweet 10 of 2014-01-01 
 Starting tweet 10 of 2014-01-03 
+Starting tweet 10 of 2014-01-01 
 Starting tweet 10 of 2014-01-02 
+Match!
+Starting tweet 11 of 2014-01-03 
 Starting tweet 11 of 2014-01-01 
 Starting tweet 11 of 2014-01-02 
-Starting tweet 11 of 2014-01-03 
+Starting tweet 12 of 2014-01-03 
 Starting tweet 12 of 2014-01-01 
 Starting tweet 12 of 2014-01-02 
-Starting tweet 12 of 2014-01-03 
+Starting tweet 13 of 2014-01-03 
+Match!
 Starting tweet 13 of 2014-01-01 
 Starting tweet 13 of 2014-01-02 
-Starting tweet 13 of 2014-01-03 
+Starting tweet 14 of 2014-01-03 
 Starting tweet 14 of 2014-01-01 
 Starting tweet 14 of 2014-01-02 
-Starting tweet 15 of 2014-01-01 
-Starting tweet 14 of 2014-01-03 
+Match!
 Starting tweet 15 of 2014-01-03 
+Starting tweet 15 of 2014-01-01 
+Match!
 Starting tweet 15 of 2014-01-02 
+Match!
+Starting tweet 16 of 2014-01-03 
 Starting tweet 16 of 2014-01-01 
 Starting tweet 16 of 2014-01-02 
-Starting tweet 16 of 2014-01-03 
-Starting tweet 17 of 2014-01-01 
 Starting tweet 17 of 2014-01-03 
+Match!
+Starting tweet 17 of 2014-01-01 
 Starting tweet 17 of 2014-01-02 
+Match!
 Starting tweet 18 of 2014-01-03 
-Starting tweet 18 of 2014-01-01 
+Match!
+Match!
 Starting tweet 18 of 2014-01-02 
+Match!
+Starting tweet 18 of 2014-01-01 
 Starting tweet 19 of 2014-01-03 
-Starting tweet 19 of 2014-01-01 
-Starting tweet 20 of 2014-01-03 
+Match!
 Starting tweet 19 of 2014-01-02 
-Starting tweet 21 of 2014-01-03 
-Starting tweet 20 of 2014-01-01 
+Starting tweet 19 of 2014-01-01 
+Match!
+Starting tweet 20 of 2014-01-03 
 Starting tweet 20 of 2014-01-02 
-Starting tweet 22 of 2014-01-03 
-Starting tweet 21 of 2014-01-01 
+Match!
+Starting tweet 20 of 2014-01-01 
+Match!
+Starting tweet 21 of 2014-01-03 
 Starting tweet 21 of 2014-01-02 
-Starting tweet 22 of 2014-01-01 
+Match!
+Starting tweet 21 of 2014-01-01 
+Starting tweet 22 of 2014-01-03 
 Starting tweet 22 of 2014-01-02 
+Starting tweet 22 of 2014-01-01 
 Starting tweet 23 of 2014-01-03 
-Starting tweet 23 of 2014-01-01 
 Starting tweet 23 of 2014-01-02 
+Starting tweet 23 of 2014-01-01 
 Starting tweet 24 of 2014-01-03 
 Starting tweet 24 of 2014-01-02 
 Starting tweet 24 of 2014-01-01 
+Match!
 Starting tweet 25 of 2014-01-03 
 Starting tweet 25 of 2014-01-02 
 Starting tweet 25 of 2014-01-01 
-Starting tweet 26 of 2014-01-03 
-Starting tweet 26 of 2014-01-02 
-Starting tweet 26 of 2014-01-01 
-Starting tweet 27 of 2014-01-03 
-Starting tweet 27 of 2014-01-02 
-Starting tweet 27 of 2014-01-01 
-Starting tweet 28 of 2014-01-03 
-Starting tweet 28 of 2014-01-02 
-Starting tweet 28 of 2014-01-01 
-Starting tweet 29 of 2014-01-03 
-Starting tweet 29 of 2014-01-01 
-Starting tweet 29 of 2014-01-02 
-Starting tweet 30 of 2014-01-03 
-Starting tweet 30 of 2014-01-01 
-Starting tweet 30 of 2014-01-02 
-Starting tweet 31 of 2014-01-03 
-Starting tweet 31 of 2014-01-02 
-Starting tweet 31 of 2014-01-01 
-Starting tweet 32 of 2014-01-02 
-Starting tweet 32 of 2014-01-03 
-Starting tweet 32 of 2014-01-01 
-Starting tweet 33 of 2014-01-02 
-Starting tweet 33 of 2014-01-03 
-Starting tweet 33 of 2014-01-01 
-Starting tweet 34 of 2014-01-02 
-Starting tweet 34 of 2014-01-03 
-Starting tweet 34 of 2014-01-01 
-Starting tweet 35 of 2014-01-02 
-Starting tweet 35 of 2014-01-03 
-Starting tweet 35 of 2014-01-01 
-Starting tweet 36 of 2014-01-02 
-Starting tweet 36 of 2014-01-01 
-Starting tweet 36 of 2014-01-03 
-Starting tweet 37 of 2014-01-01 
-Starting tweet 37 of 2014-01-02 
-Starting tweet 37 of 2014-01-03 
-Starting tweet 38 of 2014-01-01 
-Starting tweet 38 of 2014-01-02 
-Starting tweet 39 of 2014-01-01 
-Starting tweet 38 of 2014-01-03 
-Starting tweet 39 of 2014-01-02 
-Starting tweet 39 of 2014-01-03 
-Starting tweet 40 of 2014-01-01 
-Starting tweet 40 of 2014-01-02 
-Starting tweet 41 of 2014-01-01 
-Starting tweet 40 of 2014-01-03 
-Starting tweet 41 of 2014-01-02 
-Starting tweet 42 of 2014-01-01 
-Starting tweet 41 of 2014-01-03 
-Starting tweet 43 of 2014-01-01 
-Starting tweet 42 of 2014-01-02 
-Starting tweet 44 of 2014-01-01 
-Starting tweet 42 of 2014-01-03 
-Starting tweet 43 of 2014-01-02 
-Starting tweet 43 of 2014-01-03 
-Starting tweet 45 of 2014-01-01 
-Starting tweet 44 of 2014-01-02 
-Starting tweet 44 of 2014-01-03 
-Starting tweet 46 of 2014-01-01 
-Starting tweet 45 of 2014-01-02 
-Starting tweet 45 of 2014-01-03 
-Starting tweet 47 of 2014-01-01 
-Starting tweet 46 of 2014-01-02 
-Starting tweet 46 of 2014-01-03 
-Starting tweet 48 of 2014-01-01 
-Starting tweet 47 of 2014-01-02 
-Starting tweet 47 of 2014-01-03 
-Starting tweet 48 of 2014-01-02 
-Starting tweet 49 of 2014-01-01 
-Starting tweet 48 of 2014-01-03 
-Starting tweet 49 of 2014-01-02 
-Starting tweet 50 of 2014-01-01 
-Starting tweet 49 of 2014-01-03 
-Starting tweet 50 of 2014-01-02 
-Starting tweet 51 of 2014-01-01 
-Starting tweet 50 of 2014-01-03 
-Starting tweet 52 of 2014-01-01 
-Starting tweet 51 of 2014-01-02 
-Starting tweet 51 of 2014-01-03 
-Starting tweet 52 of 2014-01-02 
-Starting tweet 53 of 2014-01-01 
-Starting tweet 52 of 2014-01-03 
-Starting tweet 53 of 2014-01-02 
-Starting tweet 54 of 2014-01-01 
-Starting tweet 53 of 2014-01-03 
-Starting tweet 54 of 2014-01-02 
-Starting tweet 55 of 2014-01-01 
-Starting tweet 54 of 2014-01-03 
-Starting tweet 56 of 2014-01-01 
-Starting tweet 55 of 2014-01-02 
-Starting tweet 55 of 2014-01-03 
-Starting tweet 56 of 2014-01-02 
-Starting tweet 57 of 2014-01-01 
-Starting tweet 56 of 2014-01-03 
-Starting tweet 57 of 2014-01-02 
-Starting tweet 58 of 2014-01-01 
-Starting tweet 57 of 2014-01-03 
-Starting tweet 58 of 2014-01-02 
-Starting tweet 58 of 2014-01-03 
-Starting tweet 59 of 2014-01-01 
-Starting tweet 60 of 2014-01-01 
-Starting tweet 59 of 2014-01-03 
-Starting tweet 59 of 2014-01-02 
-Starting tweet 60 of 2014-01-03 
-Starting tweet 61 of 2014-01-01 
-Starting tweet 60 of 2014-01-02 
-Starting tweet 61 of 2014-01-03 
-Starting tweet 62 of 2014-01-01 
-Starting tweet 61 of 2014-01-02 
-Starting tweet 62 of 2014-01-03 
-Starting tweet 63 of 2014-01-01 
-Starting tweet 62 of 2014-01-02 
-Starting tweet 63 of 2014-01-03 
-Starting tweet 64 of 2014-01-01 
-Starting tweet 63 of 2014-01-02 
-Starting tweet 64 of 2014-01-03 
-Starting tweet 65 of 2014-01-01 
-Starting tweet 64 of 2014-01-02 
-Starting tweet 65 of 2014-01-03 
-Starting tweet 66 of 2014-01-01 
-Starting tweet 66 of 2014-01-03 
-Starting tweet 65 of 2014-01-02 
-2014-01-04 
-Starting tweet 1 of 2014-01-04 
-Starting tweet 67 of 2014-01-03 
-Starting tweet 66 of 2014-01-02 
-Starting tweet 2 of 2014-01-04 
-Starting tweet 67 of 2014-01-02 
-Starting tweet 68 of 2014-01-03 
-Starting tweet 3 of 2014-01-04 
-Starting tweet 68 of 2014-01-02 
-Starting tweet 69 of 2014-01-03 
-Starting tweet 4 of 2014-01-04 
-Starting tweet 70 of 2014-01-03 
-Starting tweet 69 of 2014-01-02 
-Starting tweet 5 of 2014-01-04 
-Starting tweet 71 of 2014-01-03 
-Starting tweet 70 of 2014-01-02 
-Starting tweet 72 of 2014-01-03 
-Starting tweet 6 of 2014-01-04 
-Starting tweet 71 of 2014-01-02 
-Starting tweet 73 of 2014-01-03 
-Starting tweet 7 of 2014-01-04 
-Starting tweet 74 of 2014-01-03 
-Starting tweet 72 of 2014-01-02 
-Starting tweet 75 of 2014-01-03 
-Starting tweet 8 of 2014-01-04 
-Starting tweet 73 of 2014-01-02 
-Starting tweet 76 of 2014-01-03 
-Starting tweet 9 of 2014-01-04 
-Starting tweet 74 of 2014-01-02 
-Starting tweet 77 of 2014-01-03 
-Starting tweet 75 of 2014-01-02 
-Starting tweet 10 of 2014-01-04 
-Starting tweet 78 of 2014-01-03 
-Starting tweet 76 of 2014-01-02 
-Starting tweet 11 of 2014-01-04 
-Starting tweet 77 of 2014-01-02 
-Starting tweet 79 of 2014-01-03 
-Starting tweet 80 of 2014-01-03 
-Starting tweet 12 of 2014-01-04 
-Starting tweet 78 of 2014-01-02 
-Starting tweet 79 of 2014-01-02 
-Starting tweet 13 of 2014-01-04 
-Starting tweet 81 of 2014-01-03 
-Starting tweet 80 of 2014-01-02 
-Starting tweet 14 of 2014-01-04 
-Starting tweet 82 of 2014-01-03 
-Starting tweet 81 of 2014-01-02 
-Starting tweet 15 of 2014-01-04 
-Starting tweet 83 of 2014-01-03 
-Starting tweet 82 of 2014-01-02 
-Starting tweet 84 of 2014-01-03 
-Starting tweet 16 of 2014-01-04 
-Starting tweet 83 of 2014-01-02 
-Starting tweet 17 of 2014-01-04 
-Starting tweet 85 of 2014-01-03 
-Starting tweet 84 of 2014-01-02 
-Starting tweet 18 of 2014-01-04 
-Starting tweet 86 of 2014-01-03 
-Starting tweet 85 of 2014-01-02 
-Starting tweet 19 of 2014-01-04 
-Starting tweet 87 of 2014-01-03 
-Starting tweet 88 of 2014-01-03 
-Starting tweet 86 of 2014-01-02 
-Starting tweet 20 of 2014-01-04 
-Starting tweet 89 of 2014-01-03 
-Starting tweet 87 of 2014-01-02 
-Starting tweet 21 of 2014-01-04 
-Starting tweet 90 of 2014-01-03 
-Starting tweet 88 of 2014-01-02 
-Starting tweet 22 of 2014-01-04 
-Starting tweet 91 of 2014-01-03 
-Starting tweet 89 of 2014-01-02 
-Starting tweet 23 of 2014-01-04 
-Starting tweet 92 of 2014-01-03 
-Starting tweet 24 of 2014-01-04 
-Starting tweet 90 of 2014-01-02 
-Starting tweet 93 of 2014-01-03 
-Starting tweet 91 of 2014-01-02 
-Starting tweet 25 of 2014-01-04 
-Starting tweet 94 of 2014-01-03 
-Starting tweet 92 of 2014-01-02 
-Starting tweet 26 of 2014-01-04 
-Starting tweet 95 of 2014-01-03 
-Starting tweet 93 of 2014-01-02 
-Starting tweet 96 of 2014-01-03 
-Starting tweet 27 of 2014-01-04 
-Starting tweet 94 of 2014-01-02 
-Starting tweet 28 of 2014-01-04 
-Starting tweet 97 of 2014-01-03 
-Starting tweet 95 of 2014-01-02 
-Starting tweet 29 of 2014-01-04 
-Starting tweet 98 of 2014-01-03 
-Starting tweet 96 of 2014-01-02 
-Starting tweet 30 of 2014-01-04 
-Starting tweet 99 of 2014-01-03 
-Starting tweet 97 of 2014-01-02 
-Starting tweet 31 of 2014-01-04 
-Starting tweet 100 of 2014-01-03 
-Starting tweet 98 of 2014-01-02 
-Starting tweet 32 of 2014-01-04 
-Starting tweet 101 of 2014-01-03 
-Starting tweet 99 of 2014-01-02 
-Starting tweet 33 of 2014-01-04 
-Starting tweet 102 of 2014-01-03 
-Starting tweet 100 of 2014-01-02 
-Starting tweet 34 of 2014-01-04 
-Starting tweet 103 of 2014-01-03 
-Starting tweet 101 of 2014-01-02 
-Starting tweet 35 of 2014-01-04 
-Starting tweet 104 of 2014-01-03 
-Starting tweet 102 of 2014-01-02 
-Starting tweet 36 of 2014-01-04 
-Starting tweet 105 of 2014-01-03 
-Starting tweet 103 of 2014-01-02 
-Starting tweet 37 of 2014-01-04 
-Starting tweet 106 of 2014-01-03 
-Starting tweet 104 of 2014-01-02 
-Starting tweet 38 of 2014-01-04 
-Starting tweet 107 of 2014-01-03 
-Starting tweet 105 of 2014-01-02 
-Starting tweet 108 of 2014-01-03 
-Starting tweet 39 of 2014-01-04 
-Starting tweet 106 of 2014-01-02 
-Starting tweet 107 of 2014-01-02 
-Starting tweet 109 of 2014-01-03 
-Starting tweet 40 of 2014-01-04 
-Starting tweet 108 of 2014-01-02 
-Starting tweet 110 of 2014-01-03 
-Starting tweet 41 of 2014-01-04 
-Starting tweet 109 of 2014-01-02 
-Starting tweet 110 of 2014-01-02 
-Starting tweet 111 of 2014-01-03 
-Starting tweet 42 of 2014-01-04 
-Starting tweet 111 of 2014-01-02 
-Starting tweet 112 of 2014-01-03 
-Starting tweet 43 of 2014-01-04 
-Starting tweet 113 of 2014-01-03 
-Starting tweet 112 of 2014-01-02 
-Starting tweet 44 of 2014-01-04 
-Starting tweet 114 of 2014-01-03 
-Starting tweet 113 of 2014-01-02 
-Starting tweet 45 of 2014-01-04 
-Starting tweet 115 of 2014-01-03 
-Starting tweet 114 of 2014-01-02 
-Starting tweet 46 of 2014-01-04 
-Starting tweet 115 of 2014-01-02 
-Starting tweet 116 of 2014-01-03 
-Starting tweet 47 of 2014-01-04 
-Starting tweet 117 of 2014-01-03 
-Starting tweet 116 of 2014-01-02 
-Starting tweet 118 of 2014-01-03 
-Starting tweet 48 of 2014-01-04 
-Starting tweet 117 of 2014-01-02 
-Starting tweet 119 of 2014-01-03 
-Starting tweet 118 of 2014-01-02 
-Starting tweet 49 of 2014-01-04 
-Starting tweet 120 of 2014-01-03 
-Starting tweet 119 of 2014-01-02 
-Starting tweet 50 of 2014-01-04 
-Starting tweet 121 of 2014-01-03 
-Starting tweet 120 of 2014-01-02 
-Starting tweet 51 of 2014-01-04 
-Starting tweet 122 of 2014-01-03 
-Starting tweet 121 of 2014-01-02 
-Starting tweet 52 of 2014-01-04 
-Starting tweet 123 of 2014-01-03 
-Starting tweet 122 of 2014-01-02 
-Starting tweet 53 of 2014-01-04 
-Starting tweet 123 of 2014-01-02 
-Starting tweet 124 of 2014-01-03 
-Starting tweet 54 of 2014-01-04 
-Starting tweet 124 of 2014-01-02 
-Starting tweet 125 of 2014-01-03 
-Starting tweet 55 of 2014-01-04 
-Starting tweet 125 of 2014-01-02 
-Starting tweet 126 of 2014-01-03 
-Starting tweet 56 of 2014-01-04 
-Starting tweet 126 of 2014-01-02 
-Starting tweet 57 of 2014-01-04 
-Starting tweet 127 of 2014-01-03 
-Starting tweet 127 of 2014-01-02 
-Starting tweet 58 of 2014-01-04 
-Starting tweet 128 of 2014-01-03 
-Starting tweet 128 of 2014-01-02 
-Starting tweet 129 of 2014-01-03 
-Starting tweet 59 of 2014-01-04 
-Starting tweet 129 of 2014-01-02 
-Starting tweet 60 of 2014-01-04 
-Starting tweet 130 of 2014-01-03 
-Starting tweet 130 of 2014-01-02 
-Starting tweet 61 of 2014-01-04 
-Starting tweet 131 of 2014-01-03 
-Starting tweet 131 of 2014-01-02 
-Starting tweet 132 of 2014-01-03 
-Starting tweet 62 of 2014-01-04 
-Starting tweet 63 of 2014-01-04 
-Starting tweet 132 of 2014-01-02 
-Starting tweet 133 of 2014-01-03 
-Starting tweet 64 of 2014-01-04 
-Starting tweet 133 of 2014-01-02 
-Starting tweet 134 of 2014-01-03 
-Starting tweet 65 of 2014-01-04 
-Starting tweet 135 of 2014-01-03 
-Starting tweet 134 of 2014-01-02 
-Starting tweet 66 of 2014-01-04 
-Starting tweet 136 of 2014-01-03 
-Starting tweet 135 of 2014-01-02 
-Starting tweet 137 of 2014-01-03 
-Starting tweet 67 of 2014-01-04 
-Starting tweet 136 of 2014-01-02 
-Starting tweet 138 of 2014-01-03 
-Starting tweet 68 of 2014-01-04 
-Starting tweet 137 of 2014-01-02 
-Starting tweet 69 of 2014-01-04 
-Starting tweet 138 of 2014-01-02 
-Starting tweet 70 of 2014-01-04 
-Starting tweet 139 of 2014-01-02 
-Starting tweet 71 of 2014-01-04 
-Starting tweet 140 of 2014-01-02 
-Starting tweet 72 of 2014-01-04 
-Starting tweet 141 of 2014-01-02 
-Starting tweet 73 of 2014-01-04 
-Starting tweet 74 of 2014-01-04 
-Starting tweet 75 of 2014-01-04 
-Starting tweet 76 of 2014-01-04 
-Starting tweet 77 of 2014-01-04 
-Starting tweet 78 of 2014-01-04 
-Starting tweet 79 of 2014-01-04 
-Starting tweet 80 of 2014-01-04 
-Starting tweet 81 of 2014-01-04 
-Starting tweet 82 of 2014-01-04 
-Starting tweet 83 of 2014-01-04 
-Starting tweet 84 of 2014-01-04 
-Starting tweet 85 of 2014-01-04 
-Starting tweet 86 of 2014-01-04 
+Match!