In PolMine/polmineR.misc: polmineR.misc

library(polmineR)
use("GermaParl")
library(magrittr)
library(data.table)

Q <- '"Nachhaltigkeit.*"'

Sprachliche Varianten {.smaller}

cnt <- count("GERMAPARL", query = Q, cqp = TRUE, breakdown = TRUE)
cnt[, "query" := NULL]
colnames(cnt) <- c("Wortform", "Häufigkeit", "Anteil")
DT::datatable(cnt)

Der Sprachgebrauch der Parteien {.flexbox .vcenter}

# p <- partition("GERMAPARL", role = c("government", "mp"), interjection = FALSE)

# d <- dispersion(p, query = Q, s_attribute = "party")
d <- dispersion("GERMAPARL", query = Q, s_attribute = "party")
d <- d[!party %in% c("parteilos", "")]
d <- rbind(
  d,
  data.table(
    party = "PDS/LINKE",
    count = d[party == "LINKE"][["count"]] + d[party == "PDS"][["count"]]
    )
)
d <- d[!party %in% c("PDS", "LINKE")]
setorderv(d, cols = "party")

s <- size("GERMAPARL", s_attribute = "party")
s <- rbind(
  s,
  data.table(
    party = "PDS/LINKE",
    size = s[party == "LINKE"][["size"]] + s[party == "PDS"][["size"]]
    )
)
s <- s[party %in% d[["party"]]]
setorderv(s, cols = "party")

d[, "size" := s[["size"]] ]
d[, "freq" := count / size * 100000]


colors <- c(
  "CDU" = "black", CSU = "blue", FDP = "yellow",
  SPD = "red", GRUENE = "green", "PDS/LINKE" = "pink"
  )


par(mfrow = c(1,2))
setorderv(d, cols = "freq", order = -1L)
barplot(
  height = d[["count"]], names.arg = d[["party"]], las = 2, cex.names = 0.7,
  col = colors[d[["party"]]],
  main = "absolute Häufigkeit"
  )

barplot(
  height = d[["freq"]], names.arg = d[["party"]], las = 2, cex.names = 0.7,
  col = colors[d[["party"]]],
  main = "Anteil (pro 100.000 Worte)"
  )

Kookkurrenzen | alle Fraktionen, nur Nomen und Adjektive {.smaller}

"GERMAPARL" %>%
  cooccurrences(query = Q, cqp = TRUE, p_attribute = c("pos", "word")) %>%
  subset(pos %in% c("NN", "ADJA", "ADJD")) %>%
  subset(count_window >= 5) %>%
  format() %>%
  DT::datatable()

Kookkurrenzen | CDU/CSU, nur Nomen und Adjektive {.smaller}

partition("GERMAPARL", parliamentary_group = "CDU/CSU") %>%
  cooccurrences(query = Q, cqp = TRUE, p_attribute = c("pos", "word")) %>%
  subset(pos %in% c("NN", "ADJA", "ADJD")) %>%
  subset(count_window >= 5) %>%
  format() %>%
  DT::datatable()

Kookkurrenzen | FDP, nur Nomen und Adjektive {.smaller}

partition("GERMAPARL", parliamentary_group = "FDP") %>%
  cooccurrences(query = Q, cqp = TRUE, p_attribute = c("pos", "word")) %>%
  subset(pos %in% c("NN", "ADJA", "ADJD")) %>%
  subset(count_window >= 5) %>%
  format() %>%
  DT::datatable()

Kookkurrenzen | SPD, nur Nomen und Adjektive {.smaller}

partition("GERMAPARL", parliamentary_group = "SPD") %>%
  cooccurrences(query = Q, cqp = TRUE, p_attribute = c("pos", "word")) %>%
  subset(pos %in% c("NN", "ADJA", "ADJD")) %>%
  subset(count_window >= 5) %>%
  format() %>%
  DT::datatable()

Kookkurrenzen | Bündnis 90/Die Grünen, nur Nomen und Adjektive {.smaller}

partition("GERMAPARL", parliamentary_group = c("GRUENE")) %>%
  cooccurrences(query = Q, cqp = TRUE, p_attribute = c("pos", "word")) %>%
  subset(pos %in% c("NN", "ADJA", "ADJD")) %>%
  subset(count_window >= 5) %>%
  format() %>%
  DT::datatable()

Kookkurrenzen | PDS/LINKE, nur Nomen und Adjektive {.smaller}

partition("GERMAPARL", parliamentary_group = c("PDS", "LINKE")) %>%
  cooccurrences(query = Q, cqp = TRUE, p_attribute = c("pos", "word")) %>%
  subset(pos %in% c("NN", "ADJA", "ADJD")) %>%
  subset(count_window >= 5) %>%
  format() %>%
  DT::datatable()

Konkordanzen {.smaller}

options("polmineR.pagelength" = 8)
kwic("GERMAPARL", query = Q, s_attributes = c("date", "party", "speaker"))

PolMine/polmineR.misc documentation built on Nov. 23, 2022, 9:01 p.m.