R-Code zum Blogbeitrag

#rm(list=ls(all=TRUE))
# set working directory
setwd("/Users/alexandrawuermli/Dropbox/Studium/datenjournalismus/blog")
list.of.packages <- c("dplyr", "quanteda", "foreign","ggplot2","magrittr","lubridate", "plotly", "ggthemes", "stringr", "swissparl", "tidyverse", "stats", "ggalt", "RColorBrewer", "data.table", "ggpubr", "purrr", "ggwordcloud", "zoo", "TTR", "tidyquant", "stringr", "purrr", "openxlsx", "quanteda.textstats", "deeplr", "tidyr", "tidytext", "kableExtra","magick", "webshot")


lapply(list.of.packages, require, character.only = TRUE)

theme_aw <- function(base_size = 18, 
                          base_family = "Helvetica",
                          base_line_size = base_size / 170,
                          base_rect_size = base_size / 170){
  ggplot2::theme_minimal(base_size = base_size, 
                         base_family = base_family,
                         base_line_size = base_line_size) %+replace%
    ggplot2::theme(
      plot.title = element_text(color = rgb(22, 38, 46, 
                                            maxColorValue = 250),  
                                face = "bold", hjust = 0, vjust = 0.5),
      axis.title = element_text(color = rgb(22, 38, 46, 
                                            maxColorValue = 250), 
                                size = rel(0.75)),
      axis.text = element_text(color = rgb(22, 38, 46, 
                                           maxColorValue = 250), 
                               size = rel(0.7)),  
      plot.subtitle = element_text(color = rgb(22, 38, 46, 
                                            maxColorValue = 250), hjust = 0, vjust = 0.5)
    ,
      panel.grid.minor.y  = element_line(rgb(159, 162, 178, 
                                          maxColorValue = 250),
                                      linetype = "dotted", size = rel(4)),
      legend.title = element_blank(), legend.position = "right",
      axis.ticks.y = element_blank(),
      axis.ticks = element_blank(),
      strip.background = element_blank(), 
      complete = TRUE
    )
}

Geschäfte zu Behinderung herunterladen

deutsche Geschäfte

behinderung <- get_data(
  table = "Business",
  Title = "~Behind",
  SubmissionDate = c(">1991-01-01", "<2020-12-31"), 
  Language = "DE"
)
head(behinderung)
colnames(behinderung)
length(behinderung$ID)
# 223 mit "Behind" im  Titel, manuell überprüfen!!
# write.xlsx(behinderung,"/Users/alexandrawuermli/Dropbox/Studium/datenjournalismus/blog/behind.xlsx")

# nach überprüfung nur noch 186
#behinderung2<- read.xlsx("/Users/alexandrawuermli/Dropbox/Studium/datenjournalismus/blog/behind.xlsx")

# 37 andere Themen, rausfiltern
behinderung1 <- behinderung %>%
  filter(!(BusinessShortNumber %in% c("02.1048", "02.3756", 
"06.3708","07.3679","08.3504","09.1083","09.5408","10.3130","10.5010","10.5443","11.1028","11.3398","11.4004","11.5037","12.3889","13.4089","13.5178","14.462","14.1031","14.3082","14.3316","15.3275","16.3313","16.3411","16.3529","16.3575","16.5084","17.3377","18.1075","18.3229","18.4373","18.5226","19.3333","19.3892","19.3923","19.4116","19.5500")))
behinderung1

französische Geschäfte

behinderung_f <- get_data(
  table = "Business",
  Title = "~handica",
  SubmissionDate = c(">1991-01-01", "<2020-12-31"), 
  Language = "FR"
)
head(behinderung_f)
# 171 mit "Behind" im  Titel, manuell überprüfen!!


library("openxlsx")
#write.xlsx(behinderung_f,"/Users/alexandrawuermli/Dropbox/Studium/datenjournalismus/blog/behind_f.xlsx")

behinderung1_f <- behinderung_f %>%
  filter(!(BusinessShortNumber %in% c("02.1061", "12.4097")))
behinderung1_f

# Variable unter hinzufügen, um danach zwischen den Geschäften differenzieren können
behinderung1 <- behinderung1 %>%
  mutate(unter="Behinderung")
behinderung1

behinderung1_f <- behinderung1_f %>%
  mutate(unter="Behinderung")
behinderung1_f

Geschäfte zu anderen Themen herunterladen

# deutsch
asyl<- get_data(
  table = "Business",
  Title = "~asyl",
  SubmissionDate = c(">1991-01-01", "<2020-12-31"), 
  Language = "DE"
)
head(asyl)
asyl<- asyl %>%
  mutate(unter="Asyl")
asyl
length(asyl$ID) # 799 geschäfte

# franz
asyl_f <- get_data(
  table = "Business",
  Title = "~asil",
  SubmissionDate = c(">1991-01-01", "<2020-12-31"), 
  Language = "FR"
)
head(asyl_f)
length(asyl_f$ID)
# 803 mit "asil" im  Titel

asyl_f<- asyl_f %>%
  mutate(unter="Asyl")
asyl_f

frauen <- get_data(
  table = "Business",
  Title = "~frau",
  SubmissionDate = c(">1991-01-01", "<2020-12-31"), 
  Language = "DE"
)

length(frauen$ID)
# 380 mit frau im  Titel
frauen<- frauen %>%
  mutate(unter="Frauen")
frauen

# franz
frau_f <- get_data(
  table = "Business",
  Title = "~femme",
  SubmissionDate = c(">1991-01-01", "<2020-12-31"), 
  Language = "FR"
)
head(frau_f)
length(frau_f$ID)
# 368 mit "femme" im  Titel

frau_f<- frau_f %>%
  mutate(unter="Frauen")
frau_f

klima <- get_data(
  table = "Business",
  Title = "~klima",
  SubmissionDate = c(">1991-01-01", "<2020-12-31"), 
  Language = "DE"
)


head(klima)
length(klima$ID)
klima<- klima %>%
  mutate(unter="Klima")
klima

klima_f <- get_data(
  table = "Business",
  Title = "~climat",
  SubmissionDate = c(">1991-01-01", "<2020-12-31"), 
  Language = "FR"
)
head(klima_f)
length(klima_f$ID)
# 350 mit "femme" im  Titel
klima_f<- klima_f %>%
  mutate(unter="Klima")
klima_f

flücht <- get_data(
  table = "Business",
  Title = "~flücht",
  SubmissionDate = c(">1991-01-01", "<2020-12-31"), 
  Language = "DE"
)


head(flücht)
length(flücht$ID) # 335

flücht<- flücht %>%
  mutate(unter="Geflüchtete")
flücht

flucht_f <- get_data(
  table = "Business",
  Title = "~refug",
  SubmissionDate = c(">1991-01-01", "<2020-12-31"), 
  Language = "FR"
)
head(flucht_f)
length(flucht_f$ID)
# 322 

flucht_f<- flucht_f %>%
  mutate(unter="Geflüchtete")
flucht_f

ahv <- get_data(
  table = "Business",
  Title = "~ahv",
  SubmissionDate = c(">1991-01-01", "<2020-12-31"), 
  Language = "DE"
)


head(ahv)
length(ahv$ID) #334
ahv<- ahv %>%
  mutate(unter="AHV")
ahv

ahv_f <- get_data(
  table = "Business",
  Title = "~avs",
  SubmissionDate = c(">1991-01-01", "<2020-12-31"), 
  Language = "FR"
)
head(ahv_f)
length(ahv_f$ID)
# 331

ahv_f<- ahv_f %>%
  mutate(unter="AHV")
ahv_f

bildung <- get_data(
  table = "Business",
  Title = "~bildung",
  SubmissionDate = c(">1991-01-01", "<2020-12-31"), 
  Language = "DE"
)


head(bildung)
length(bildung$ID)

bildung<- bildung %>%
  mutate(unter="Bildung")
bildung


bildung_f <- get_data(
  table = "Business",
  Title = "~formation",
  SubmissionDate = c(">1991-01-01", "<2020-12-31"), 
  Language = "FR"
)
head(bildung_f)
length(bildung_f$ID)
# 1208

bildung_f<- bildung_f %>%
  mutate(unter="Bildung")
bildung_f

covid <- get_data(
  table = "Business",
  Title = "~covid",
  SubmissionDate = c(">1991-01-01", "<2020-12-31"), 
  Language = "DE"
)


head(covid)
length(covid$ID)

covid<- covid %>%
  mutate(unter="Covid-19")
covid

covid_f <- get_data(
  table = "Business",
  Title = "~covid",
  SubmissionDate = c(">1991-01-01", "<2020-12-31"), 
  Language = "FR"
)
head(covid_f)
length(covid_f$ID)
# 331

covid_f<- covid_f %>%
  mutate(unter="Covid-19")
covid_f

armee <- get_data(
  table = "Business",
  Title = "~armee",
  SubmissionDate = c(">1991-01-01", "<2020-12-31"), 
  Language = "DE"
)


head(armee)
length(armee$ID)
armee<- armee %>%
  mutate(unter="Armee")
armee

armee_f <- get_data(
  table = "Business",
  Title = "~armée",
  SubmissionDate = c(">1991-01-01", "<2020-12-31"), 
  Language = "FR"
)
head(armee_f)
length(armee_f$ID)
# 531

armee_f<- armee_f %>%
  mutate(unter="Armee")
armee_f

beh<- rbind(behinderung1,behinderung1_f)
save(beh, file=paste("behinderung_final",".Rda", sep=""))

load("behinderung_final.Rda")
data1<- beh %>%
  mutate(date= year(SubmissionDate)) %>%
  select(date, Title) %>%
    group_by(date) %>%
  summarise(n = n()) %>%
rename(
    n_behin = n )

Abbildung 1

#column-line-timeline
plot3<-ggplot(data1, aes(x=date, y = n_behin)) + 
  geom_col(aes( fill= "#3d5a80" )) +
geom_ma(ma_fun = SMA, n = 7, aes(color = "#abd9e9")) +
   scale_fill_manual(values=c("#3d5a80"), labels = c("absolute Anzahl", "gleitender 5-Jahresdurchschnitt"))+
   scale_color_manual(values=c("#abd9e9"), labels = c( "gleitender 7-Jahresdurchschnitt"))+
  annotate(
    geom = "curve", x = 2016, y = 24, xend = 2014, yend = 10, 
    curvature = -.1, arrow = arrow(length = unit(2, "mm"))
  ) +
  annotate(geom = "text", x = 2012, y = 28, 
           label = "Ratifizierung der \nBehindertenrechtskonvention (BRK),\nApril 2014",
           family = "sans",
           size = 3,
           hjust = "left") +
  annotate(
    geom = "curve", x = 2002, y = 28, xend = 2004, yend = 23, 
    curvature = -.2, arrow = arrow(length = unit(2, "mm"))
  ) +
  annotate(geom = "text", x = 1999, y = 30, 
         label = "Inkrafttreten des Behinderten- \ngleichstellungsgesetz (BehiG), \nJanuar 2004 ",
         family = "sans",
         size = 3,
         hjust = "left") +
  
  
  annotate(
    geom = "curve", x = 2018, y = 38, xend = 2019, yend = 42, 
    curvature = -.2, arrow = arrow(length = unit(2, "mm"))
  ) +
  annotate(geom = "text", x = 2015, y = 37, 
         label = "Beginn der Corona-Krise",
         family = "sans",
         size = 3,
         hjust = "left") +
  
    labs(title = "Abb. 1: Leicht steigende Tendenz beim Thema Behinderung", subtitle= "Anzahl der eingereichten Geschäfte, in denen es um Behinderung geht", caption = "Daten: parlament.ch", x = "Jahr",y = "Anzahl Geschäfte") +
  theme_aw() +theme(legend.position = "bottom")

ggsave(plot3, file = "abb.1.png", width = 12, height = 7)

plot3

alle Datensätze zu einem grossen zusammenfügen

# datensätze zusammensetzen
data2<-rbind(behinderung1,behinderung1_f, armee, armee_f,frauen,frau_f,asyl, asyl_f, klima, klima_f, flücht,flucht_f, ahv, ahv_f, covid,covid_f, bildung, bildung_f)
save(data2, file=paste("datensatz_final",".Rda", sep=""))

load("datensatz_final.Rda")

Abbildung 2

plot2<-data2%>%
  select(Title, unter)%>%
group_by(unter)%>%
  summarise(n = n())%>%
arrange(desc(n)) %>% 
ggplot(aes(x=fct_reorder(unter, n), y = n)) +
  geom_bar(stat = "identity", fill = "#3d5a80", color = "white")+
  geom_text(
      aes(
        y = n,
        label = n
      ), 
      position = position_stack(vjust = 0.5),
      color = "white",
      family = "Helvetica",
      size = 5
    ) +
labs(title = "Abb. 2: Im Vergleich hinkt das Thema Behinderung hinterher", subtitle= "Anzahl der eingereichten Geschäfte im Vergleich", caption = "Daten: parlament.ch", x = "Themen",y = "Anzahl Geschäfte") +
  theme_aw() 

`summarise()` ungrouping output (override with `.groups` argument)

ggsave(plot2, file = "abb.2.png", width = 12, height = 7)

plot2

Umbenennen

data2$BusinessTypeName[data2$BusinessTypeName == "Question ordinaire"] <- "Einfache Anfrage"
data2$BusinessTypeName[data2$BusinessTypeName == "Recommandation"] <- "Empfehlung"
data2$BusinessTypeName[data2$BusinessTypeName == "Heure des questions. Question"] <- "Fragestunde. Frage"
data2$BusinessTypeName[data2$BusinessTypeName == "Initiative déposée par un canton"] <- "Standesinitiative"
data2$BusinessTypeName[data2$BusinessTypeName == "Question ordinaire urgente"] <- "Dringliche Einfache Anfrage"
data2$BusinessTypeName[data2$BusinessTypeName == "Question urgente"] <- "Dringliche Anfrage"
data2$BusinessTypeName[data2$BusinessTypeName == "Objet du Parlement"    ] <- "Geschäft des Parlaments"
data2$BusinessTypeName[data2$BusinessTypeName == "Initiative parlementaire"   ] <- "Parlamentarische Initiative"
data2$BusinessTypeName[data2$BusinessTypeName == "Pétition"   ] <- "Petition"
data2$BusinessTypeName[data2$BusinessTypeName == "Objet du Conseil fédéral"  ] <- "Geschäft des Bundesrates"
data2$BusinessTypeName[data2$BusinessTypeName == "Question"  ] <- "Anfrage"
data2$BusinessTypeName[data2$BusinessTypeName == "Interpellation urgente"] <- "Dringliche Interpellation"

data3<-data2 %>%
  mutate(date = year(SubmissionDate)) %>%
  select(date,unter, BusinessTypeName) %>%
    group_by(unter, BusinessTypeName) %>%
  summarise(Percentage=n()) %>% 
  na.omit() %>%
  group_by(unter) %>% 
  mutate(pcent=Percentage/sum(Percentage)*100)%>% 
filter(pcent>1)

Abbildung 3

# Businness Type Behinderung
plot4<-data3 %>%
  ungroup %>%
    mutate(unter = as.factor(unter),
           BusinessTypeName = reorder_within(BusinessTypeName, pcent, unter)) %>%
ggplot(aes(BusinessTypeName, pcent,fill= unter)) +
  geom_point(size=2, color=ifelse(data3$BusinessTypeName %in% c("Petition", "Parlamentarische Initiative", "Geschäft des Bundesrates" , "Motion" , "Standesinitiative"), "#abd9e9", "#3d5a80"), show.legend = FALSE) + 
   geom_segment(aes(x=BusinessTypeName, xend=BusinessTypeName, y=0, yend=pcent), color="grey")  +
  geom_text(aes(label = sprintf("%.0f%%", pcent)), vjust = 0.5, hjust = -0.5, size = 3, color = "black") +
    labs(title = "Abb. 3: Es wird vor allem geredet..", subtitle= "Vergleich zwischen der Art der Geschäfte (in %)", caption = "Daten: parlament.ch", x = "",y = "") +
  theme_light() +
  theme(
    panel.grid.major.x = element_blank(),
    panel.border = element_blank(),
    axis.ticks.x = element_blank()) +
facet_wrap(~ unter, scales = "free_y")+
  scale_x_reordered() +
    scale_y_continuous(limits=c(0, 60),labels = scales::label_percent(scale=1)) +
    coord_flip() +
    theme_aw() 

ggsave(plot4, file = "abb.3.png", width = 15, height = 12)

plot4 

um die Texte zu den Geschäften herunterzuladen, brauchen wir zuerst die BusinessShortNumber.

df<-behinderung1 %>% 
  select(BusinessShortNumber, Title) 
df
# 186 Geschäfte deutsch

df1<-behinderung1_f %>% 
  select(BusinessShortNumber, Title) 
df1
# 169 Geschäfte franz

Texte zu Geschäfte herunterladen

deutsch

beh_d <- swissparl::get_data(
  table = "SubjectBusiness",
  BusinessShortNumber = df$BusinessShortNumber,
  Language = "DE"
  )
beh_d

transcripts_beh_d<- swissparl::get_data(
  table = "Transcript", 
  IdSubject = as.numeric(beh_d$IdSubject),
  Language = "DE",
  Type = 1,
  LanguageOfText = "DE"
  )
transcripts_beh_d
length(transcripts_beh_d$Text)
# 326 Texte BR, NR, ST


# zur sicherheit speichern
#saveRDS(transcripts_beh_d, file = "transcripts_beh_deutsch.RDS") 
#readRDS("transcripts_beh.RDS") 

Texte zu Geschäfte herunterladen

franz

beh_f <- swissparl::get_data(
  table = "SubjectBusiness",
  BusinessShortNumber = df1$BusinessShortNumber,
  Language = "FR"
  )
beh_f

transcripts_beh_f<- swissparl::get_data(
  table = "Transcript", 
  IdSubject = as.numeric(beh_f$IdSubject),
  Language = "FR",
  Type = 1,
  LanguageOfText = "FR"
  )
transcripts_beh_f
length(transcripts_beh_f$Text)
# 129

# zur sicherheit speichern
#saveRDS(transcripts_beh_f, file = "transcripts_beh_franz.RDS") 
#readRDS("transcripts_beh_franz.RDS") 

Zusammenfügen der Texte

transcripts_beh<- rbind(transcripts_beh_f, transcripts_beh_d)
#saveRDS(transcripts_beh, file = "transcripts_final.RDS") 

transcripts_beh<-readRDS("transcripts_final.RDS") 

Daten vorbereiten

alle_texte <- transcripts_beh %>%
  mutate(Text2 = clean_text(Text, keep_round_brackets = F))
head(alle_texte)

alle_texte %<>%
  filter(!SpeakerLastName == "leer") %>% 
  filter(!SpeakerFunction %in% c("1VP-F", "1VP-M", "2VP-F", "2VP-M", "AP-M", "P-F", "P-M")) %>% 
  filter(!Function %in% c("1VP-M", "2VP-M", "P-F", "p-m", "P-m", "P-M", "P-MM")) %>%
  filter(!SpeakerFullName == "Thurnherr Walter") %>%
  filter(!CouncilName %in% c("Bundesrat", "Conseil fédéral")) 
length(alle_texte$Text2)

Fraktionen umbenennen

alle_texte$ParlGroupName[alle_texte$ParlGroupName == "Groupe de l'Union démocratique du Centre"] <- "Fraktion der Schweizerischen Volkspartei"
alle_texte$ParlGroupName[alle_texte$ParlGroupName == "Groupe radical-libéral"] <- "FDP Fraktion"
alle_texte$ParlGroupName[alle_texte$ParlGroupName == "Groupe libéral-radical"] <- "FDP Fraktion"
alle_texte$ParlGroupName[alle_texte$ParlGroupName == "Groupe PDC/PEV/PVL"] <- "Die Mitte-Fraktion. Die Mitte. EVP."
alle_texte$ParlGroupName[alle_texte$ParlGroupName == "Groupe PDC/PEV/PVL"] <- "Die Mitte-Fraktion. Die Mitte. EVP."

alle_texte$ParlGroupName[alle_texte$ParlGroupName == "Le Groupe du Centre. Le Centre. PEV."] <- "Die Mitte-Fraktion. Die Mitte. EVP."
alle_texte$ParlGroupName[alle_texte$ParlGroupName == "Christlichdemokratische Fraktion"    ] <- "Die Mitte-Fraktion. Die Mitte. EVP."
alle_texte$ParlGroupName[alle_texte$ParlGroupName == "Groupe PDC"   ] <- "Die Mitte-Fraktion. Die Mitte. EVP."
alle_texte$ParlGroupName[alle_texte$ParlGroupName == "CVP-Fraktion"   ] <- "Die Mitte-Fraktion. Die Mitte. EVP."
alle_texte$ParlGroupName[alle_texte$ParlGroupName == "Groupe PDC-PEV"  ] <- "Die Mitte-Fraktion. Die Mitte. EVP."
alle_texte$ParlGroupName[alle_texte$ParlGroupName == "Groupe PDC/PEV/PVL"  ] <- "Die Mitte-Fraktion. Die Mitte. EVP."
alle_texte$ParlGroupName[alle_texte$ParlGroupName == "Groupe démocrate-chrétien"] <- "Die Mitte-Fraktion. Die Mitte. EVP."
alle_texte$ParlGroupName[alle_texte$ParlGroupName == "Groupe socialiste"] <- "Sozialdemokratische Fraktion"
alle_texte$ParlGroupName[alle_texte$ParlGroupName == "Groupe des VERT-E-S"] <- "Grüne Fraktion"
alle_texte$ParlGroupName[alle_texte$ParlGroupName == "Groupe écologiste"] <- "Grüne Fraktion"
alle_texte$ParlGroupName[alle_texte$ParlGroupName == "Fraktion CVP-EVP" ] <- "Die Mitte-Fraktion. Die Mitte. EVP."
alle_texte$ParlGroupName[alle_texte$ParlGroupName == "CVP-EVP-BDP Fraktion"  ] <- "Die Mitte-Fraktion. Die Mitte. EVP."
alle_texte$ParlGroupName[alle_texte$ParlGroupName == "Fraktion CVP/EVP/glp" ] <- "Die Mitte-Fraktion. Die Mitte. EVP."
alle_texte$ParlGroupName[alle_texte$ParlGroupName == "Evangelische und Unabhängige Fraktion" ] <- "Die Mitte-Fraktion. Die Mitte. EVP."
alle_texte$ParlGroupName[alle_texte$ParlGroupName == "Fraktion BD" ] <- "Die Mitte-Fraktion. Die Mitte. EVP."

alle_texte$ParlGroupName[alle_texte$ParlGroupName == "Liberale Fraktion"] <-  "FDP Fraktion"
alle_texte$ParlGroupName[alle_texte$ParlGroupName == "FDP-Liberale Fraktion"] <-  "FDP Fraktion"
alle_texte$ParlGroupName[alle_texte$ParlGroupName == "Freisinnig-demokratische Fraktion"] <-  "FDP Fraktion"
alle_texte$ParlGroupName[alle_texte$ParlGroupName == "Fraktion der Schweizerischen Volkspartei"] <-  "SVP Fraktion"
alle_texte$ParlGroupName[alle_texte$ParlGroupName == "Sozialdemokratische Fraktion"] <-  "SP Fraktion"
unique(alle_texte$ParlGroupName)

stopwords("de")
stopwords("fr")[1:150]
mystopwords <- c( "Behinderung", "Behinderungen","initiative", "gesetzes", "neuen", "punkt", "gar", "sollten" ,"stellen", "behinderte", "behinderten", "menschen", "frage", "sei", "wichtig", "folgen", "Kommissionsmehrheit", "sieht","monate", "ausgeht","person","gesetz","sollen","beiden","weit","gehen","besteht","möglich", "genau","block", "kollege"," möglichst", "müssen","mehr","bitte","geht","sich","tun" ,"gibt","darf", "möchte","für", "dass", "letzte", "letzt", "letzter", "letztes", "vorletzte", "ausserdem", "bereits", "erster", "zweiter", "dritter", "erste", "zweite", "dritte", "erstens", "zweitens", "drittens", "grundsätzlich", "bereich", "kantone",
                 "soeben", "nebst", "neben", "kurz" , "immer" , "selten" , "nie" , "manchmal" , "oft" , "zwischendurch" , "seitdem", "erst",
                 "ja", "z.B.", "der", "die", "das", "ab", "deshalb", "dafür", "heute", "sofort", "danach", "schon", "bald", "haben", "hat", "habt", "hatten",
                 "wäre", "viele", "vieles", "vielen", "vielem", "einfach", "so", "davon", "vielleicht", "weil", "aufgrund", "gerade", "wurden", "gegen", "wider",
                 "wurde", "wurdest", "deshalb", "dafür", "darum", "dagegen", "selbst", "selber", "weitere", "weiteres", "weiterer", "ganz", "eben", "direkt", 
                 "notre", "avons", "avez", "ont", "la", "le", "de", "I", "dovra", "Conseil", "fédéral", "Bundesrat", "Bundesrates", "Bundesräte", "confédération",
                 "Nationalrat", "Ständerat", "für", "dass", "commission", "Kommission", "letzte", "letzt", "letzter", "letztes", "vorletzte", "Minderheit", "Mehrheit",
                 "Vorlage", "Motion", "Artikel", "Absatz", "soeben", "nebst", "neben", "gleich", "ja", "nein", "z.B.", "seit", "rund", "klar", "Prozent",
                 "ausserdem", "bereits", "Antrag", "wirklich", "auch", "ebenfalls", "der", "die", "das", "ab", "deshalb", "dafür", "heute", "sofort", 
                 "daher", "natürlich", "insbesondere", "zudem", "darauf", "nämlich", "dabei","eigentlich", "schon", "sowie", "gemäss", "gleichzeitig", 
                 "Beispiel", "beim", "letzten","worden","insgesamt", "deshalb", "dafür", "darum", "dagegen", "selbst", "selber", "weitere",    "weiteres",
                 "weiterer", "ganz", "entsprechend", "entsprechendes","entsprechender", "notre", "avons", "avez", "ont", "domain", "certain", "majorité", "minorité", 
                 "la", "le", "de", "I", "dovra", "donc", "c'est","plus", "puisque", "pendant", "ni", "fair", "fait", "être", "tout", "tous", "toutes", "toute",
                 "aussi", "enfaite", "souvent", "aujourd'hui", "en effet", "lorsque", "donc", "l'article", "n'est", "c'est", "elle", "elles", "contre","cent",
                 "ussa", "nostro", "c'è", "che", "per", "a", "e", "di", "perche", "z.B", "ein", "zwei", "drei", "vier", "unserer", "unsere", "unser","déjà",
                 "b", "z", "i", "ii", "d'un", "d'autre", "1a", "1", "2", "3", "4", "h", "a", "cas", "mais", "un", "deux", "trois", "quatre", "cinq", "faut",
                 "d'une", "sagen", "gesagt", "très", "sehr", "herr", "frau", "monsieur", "madame", "également", "comme", "évidemment", "déjà", "l'on",
                 "où", "soutien", "concernant", "oui", "non", "si", "qu'il", "qu'elle", "qu'ils", "qu'elles", "encore", "ainsi", "entre", "lors", "dont", 
                 "2019", "2020", "2021", "demande", "question", "pour", "cent", "personnes", "handicapées", "bien", "loi" , "handicapés", "l'initiative", "peut", "handicap", "suisse", "alinéa", "faire")

#corpus machen
beh_corpus <- corpus(
    alle_texte$Text2,
    docnames = alle_texte$ID,
    docvars = alle_texte %>% select(-Text, -Text2)
    ) 

beh_corpus

#Tokenisieren
beh_toks <- tokens(beh_corpus,
                       remove_punct=T,
                       remove_numbers=T, remove_symbols=T)

beh_corpus_final <- beh_toks %>%
  tokens_remove(stopwords('de')) %>%
  tokens_remove(stopwords('fr')) %>%
  tokens_remove(mystopwords)

# Create dfm
beh_dfm <- dfm(beh_corpus_final,
               tolower = T,) 
beh_dfm 

Welche Parteien/Fraktionen/ParlamentarierInnen beschäftigen sich am meisten mit diesem Thema Behinderung?

#ParlamentarierInnen 
alle_texte$CouncilName[alle_texte$CouncilName == "Conseil des Etats"] <-  "Ständerat"
alle_texte$CouncilName[alle_texte$CouncilName == "Conseil national"] <-  "Nationalrat"
unique(alle_texte$CouncilName)


most_speeches_speaker <- alle_texte %>%
  group_by(SpeakerFullName, ParlGroupName, CouncilName ) %>%
   summarise(n=n()) %>%
  arrange(desc(n))%>%
filter(n>3)

Tabelle erstellen

SpeakerFullName <- as.character(unique(most_speeches_speaker$SpeakerFullName))
ParlGroupName <- as.character(unique(most_speeches_speaker$ParlGroupName))
Council <- as.character(unique(most_speeches_speaker$CouncilName))
Reden <- as.character(most_speeches_speaker$n)


table <- as.data.frame(cbind(SpeakerFullName, ParlGroupName, Council, Reden))
colnames(table) <- c("ParlamentarierIn", "Fraktion","Rat", "Anzahl Reden")
  table %>%
kable(row.names = F) %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed", "responsive"), full_width = T, fixed_thead = T, font_size = 13) %>%
 save_kable(file = "table_2.html",
             zoom = 1.5)
  

brändli<-alle_texte%>%
  filter(SpeakerLastName =="Brändli")
brändli

Abbildung 4

p1 <- ggplot(data = most_speeches_speaker) +
    aes(x = fct_reorder(SpeakerFullName, n), y = n) +
    geom_point(size=2, color="#3d5a80", show.legend = FALSE) + 
   geom_segment(aes(x=SpeakerFullName, xend=SpeakerFullName, y=0, yend=n), color="grey")+
    labs(title = "Abb. 4: Ch. Brändli von der SVP hat am Meisten gesprochen", subtitle = "Anzahl Reden im Vergleich", caption = "Daten: parlament.ch", y = "Anzahl", x = "") +
  coord_flip() +
theme_light() +
theme_aw() 


ggsave(p1, file = "abb.7.png", width = 11, height = 7)

partei <- alle_texte %>%
  group_by(ParlGroupName) %>%
  summarise(n=n()) %>%
  mutate(pcent=n/sum(n)*100)%>% 
  arrange(desc(n))

Abbildung 4

#plot Redeanteil nach parteien
p2 <- ggplot(data = partei) +
    aes(x = fct_reorder(ParlGroupName, pcent), y = pcent) +
    geom_point(size=6, color=c( "#dd0e0e", "#ff9100",  "#3a8bc1",  "#0a7228", 
                                "#66cc00", "#a0a000")) + 
   geom_segment(aes(x=ParlGroupName, xend=ParlGroupName, y=0, yend=pcent), color="grey")+
   geom_text(aes(label = sprintf("%.0f%%", pcent)), vjust = -1.7, hjust = 0.5, size = 3, color = "black")+
    labs(title = "Abb. 4: Ingesamt spricht die SP-Fraktion am häufigsten über Behinderung", subtitle = "Anzahl Wortbeträge im Vergleich (in %)", caption = "Daten: parlament.ch", y = "Prozent", x = "") +
  coord_flip() +
  scale_y_continuous(limits=c(0, 30),labels = scales::label_percent(scale=1))+
theme_light() +
theme_aw() +
  theme(axis.text.x = element_text(size=10))

p2
ggsave(p2, file = "abb.6.png", width = 12, height = 7)

textplot_wordcloud(beh_dfm, max_words = 50, min_count=5,
                                    color = RColorBrewer::brewer.pal(9, "GnBu")[3:8], random_order = FALSE,random_color=FALSE, rotation=0.25)

beh_freq <- textstat_frequency(beh_dfm , n=50)

Abbildung 5

png("Abb.5.png",width=10,height=12,units = "in",res=300)
                 textplot_wordcloud(beh_dfm, max_words = 50, min_count=5,color = RColorBrewer::brewer.pal(9, "GnBu")[3:8], random_order = FALSE,random_color=FALSE, rotation=0.25)
title("Abb. 5: Gleichstellung ist das meistverwendete Wort",cex.main = 1.8,
       font.main= 2,col.main="#05032d",adj=0)
mtext("Daten: parlament.ch", cex=1,side=1, line=3, adj = 1, col="grey30")
mtext("Die am häufigsten genannten Begriffe nach Anzahl Erwähnungen", cex=1.5,side=3, adj = 0, col="#05032d")

Kontext überprüfen

kwic(beh_corpus_final, phrase("integration")) 
kwic(beh_corpus_final, phrase("gleichstellung")) 
kwic(beh_corpus_final, phrase("bauten")) 
kwic(beh_corpus_final, phrase("kinder")) 
kwic(beh_corpus_final, phrase("franken")) 
kwic(beh_corpus_final, phrase("kosten"))