R
!!!relativeFrequency 相対頻度について
{{outline}}

!!サンプルデータ　１００万語
http://www.thegrammarlab.com/?nor-portfolio=1000000-word-sample-corpora
http://micusp.elicorpora.info/
MICUSP Sample
http://www.thegrammarlab.com/?nor-portfolio=1000000-word-sample-corpora#
{{pre
#　サンプルデータを読み込む
install.packages("readtext", dependencies = T)
library(readtext)

sample100 <- readtext(choose.files())

sample100
str(sample100)

install.packages("quanteda")
library(quanteda)


sample100.corpus <- corpus(sample100)

summary(sample100.corpus)

sample100.dfm <- dfm(sample100.corpus)

summary(sample100.dfm)


sample100.tokens <- tokens(sample100orpus)

#サンプリング

tokens_chunk(sample100.tokens, 100)
}}
!!全体の高頻度語
{{pre
> topfeatures(sample100.dfm)
  the    of    to   and    in     a    is  that   for    as 
68659 38144 28852 28657 22834 19779 15292 14552  9630  9208 

> topfeatures(sample100.dfm, 100)
      the        of        to       and        in         a        is      that       for        as      this        be       are      with 
    68659     38144     28852     28657     22834     19779     15292     14552      9630      9208      7612      7154      6860      6814 
        s        it        on       not        by       was      from        an     their      have        or     which        at         i 
     6396      6227      5998      5401      5374      4712      4497      4194      3932      3820      3804      3415      3314      3231 
     they       his      were        we       can      more       one     these      will        he       has       but      also     would 
     3171      3123      2953      2898      2866      2837      2833      2817      2802      2645      2600      2541      2425      2424 
    other       all     there       her   between      when      than        if       two      only   because     about      such       may 
     2067      2056      1960      1867      1863      1832      1795      1775      1695      1675      1666      1634      1584      1534 
  however      time      been      what       its      into       who        so  students      each      both     first    social      most 
     1515      1505      1478      1475      1467      1465      1407      1384      1379      1367      1355      1345      1340      1328 
      how       had     while      some        do    people       she   through      used        no     could      does      many      them 
     1308      1300      1258      1258      1248      1225      1220      1206      1202      1198      1185      1179      1170      1167 
different      then       our        my     women       use       out     state     being      well      data     study    system    should 
     1145      1083      1077      1049      1023      1016      1015      1000       985       973       971       961       943       938 
     even     where 
      930       906 
}}