RMeCab RCaBoCha p. 2/29 RMeCab R MeCab R MeCab

Size: px

Start display at page:

Download "RMeCab RCaBoCha p. 2/29 RMeCab R MeCab R MeCab"

としはるますはら
5 years ago
Views:

1 RMeCab RCaBoCha p. 1/29 RMeCab RCaBoCha

2 RMeCab RCaBoCha p. 2/29 RMeCab R MeCab R MeCab

3 RMeCab RCaBoCha p. 2/29 RMeCab R MeCab R MeCab

4 RMeCab RCaBoCha p. 3/29 footnote2 RMeCab R MeCab R MeCab Google Windows MeCab R RMeCab MeCab MeCab

5 RMeCab RCaBoCha p. 4/29 mecab # EOS,,*,*,,,,,,,,*,*,*,,,,,,*,*,*,,,,,*,*,,,,,,*,*,*,,,,,,,*,*,*,*,,,

6 RMeCab RCaBoCha p. 4/29 mecab # EOS,,*,*,,,,,,,,*,*,*,,,,,,*,*,*,,,,,*,*,,,,,,*,*,*,,,,,,,*,*,*,*,,, MeCab

7 footnote3 MeCab MeCab Excel R R MeCab 2 RMeCab R RMeCab RMeCab RCaBoCha p. 5/29

8 RMeCab RCaBoCha p. 6/29 RMeCab R ( ) C++ MeCab (libmecab.{so,dll})

9 RMeCab RCaBoCha p. 6/29 RMeCab R ( ) C++ MeCab (libmecab.{so,dll}) 2008 CRAN

10 RMeCab RCaBoCha p. 6/29 RMeCab R ( ) C++ MeCab (libmecab.{so,dll}) 2008 CRAN ver. RMeCab_0.91 (.mecabrc )

11 RMeCab RCaBoCha p. 6/29 RMeCab R ( ) C++ MeCab (libmecab.{so,dll}) 2008 CRAN ver. RMeCab_0.91 (.mecabrc ) R

12 RMeCab RCaBoCha p. 7/29 footnote4 RMeCab R R R RMeCab C++ MeCab R CRAN R RMeCab RMeCab

13 RMeCab RCaBoCha p. 8/29 RMeCab R MeCab > library(rmecab) > unlist (RMeCabC(" ", 1) ) " " " " " " " " " " " " " " > res<-rmecabtext("yukiguni.txt") file = yukiguni.txt > res [[1]] [1] " " " " " " "*" "*" [6] "*" "*" " " " " " " [[2]] [1] " " " " " " " " "*" [6] "*" "*" " " " " " "

14 RMeCab RCaBoCha p. 9/29 footnote5 RMeCab RMeCabC R 2 1 RMeCabText MeCab MeCab R RMeCab

15 RMeCab RCaBoCha p. 10/29 MeCab > docdf("doc",pos = c(" "," "),type = 1) TERM POS1 POS2 doc1 doc2 doc

16 RMeCab RCaBoCha p. 10/29 MeCab > docdf("doc",pos = c(" "," "),type = 1) > docdf("doc") TERM POS1 POS2 doc1 doc2 doc3 Ngram doc1 doc2 doc

17 RMeCab RCaBoCha p. 10/29 MeCab > docdf("doc",pos = c(" "," "),type = 1) > docdf("doc") TERM POS1 POS2 doc1 doc2 doc3 Ngram doc1 doc2 doc > docdf("doc", pos = c(" "," "), co = 1) TERM POS1 POS

18 RMeCab RCaBoCha p. 10/29 MeCab > docdf("doc",pos = c(" "," "),type = 1) > docdf("doc") TERM POS1 POS2 doc1 doc2 doc3 Ngram doc1 doc2 doc > docdf("doc", pos = c(" "," "), co = 1) TERM POS1 POS node > collocate("kumo", node = " ", span =3) Term Before After Span Total MORPHEMS TOKENS

19 RMeCab RCaBoCha p. 11/29 footnote6 RMeCab MeCab R docdf 3 doc docdf mecab RMeCab Before node After Span Total

20 RMeCab RCaBoCha p. 12/29 Ngram bigram > docdf("yukiguni", type = 1,N = 2) TERM POS1 POS2 yukiguni *-* *- 3

21 RMeCab RCaBoCha p. 12/29 Ngram bigram > docdf("yukiguni", type = 1,N = 2) TERM POS1 POS2 yukiguni *-* *- 3 trigram > docdf("yukiguni", N = 3) Ngram yukiguni

22 RMeCab RCaBoCha p. 13/29 footnote NGram N

23 RMeCab RCaBoCha p. 14/29 Ngram 2 trigram > docdf("morikita", pos = c(" "," "), type = 1, N =3) - - TERM POS1 POS2 doc1 doc2 doc

24 RMeCab RCaBoCha p. 14/29 Ngram 2 trigram > docdf("morikita", pos = c(" "," "), type = 1, N =3) TERM POS1 POS2 doc1 doc2 doc trigram > docdf("morikita", pos = c(" "," "), type = 1, N =3, ndf = 1 ) N1 N2 N3 POS1 POS2 doc1 doc2 doc

25 RMeCab RCaBoCha p. 15/29 footnote Ngram RMeCab Ngram trigram R

26 RMeCab RCaBoCha p. 16/29 hato <- docdf("hato.txt", type = 1, N = 2, ndf = 1) hato <- hato[hato$hato.thatot > 2,] hato <- hato[,c(1,2,5)] library (igraph) hato.g <- graph.data.frame(hato) E(hato.g)$weight <- hato[,3] tkplot(hato.g, vertex.label = V(hato.g)$name, vertex.size = 1, edge.label = E(hato.g)$weight)

27 RMeCab RCaBoCha p. 17/29 footnote R igraph

28 RMeCab RCaBoCha p. 18/

29 RMeCab RCaBoCha p. 19/29

30 term RMeCab RCaBoCha p. 20/29

31 RMeCab RCaBoCha p. 21/29 RMeCab (csv ) RMeCabDF docmatrix, docdf Ngram docngramdf docdf RMeCabText MeCabFreq, docdf docmatrix docmatrix2, docdf Ngram Ngram NgramDF NgramDF2, docngram, docngram2, docdf

32 RMeCab RCaBoCha p. 21/29 RMeCab (csv ) RMeCabDF docmatrix, docdf Ngram docngramdf docdf RMeCabText MeCabFreq, docdf docmatrix docmatrix2, docdf Ngram Ngram NgramDF NgramDF2, docngram, docngram2, docdf docdf

33 RMeCab RCaBoCha p. 22/29 footnote RMeCab tfidf RMeCab CSV docdf.

34 RMeCab RCaBoCha p. 23/29 RMeCab RCaBoCha

35 RMeCab RCaBoCha p. 23/29 RMeCab RCaBoCha

36 RMeCab RCaBoCha p. 23/29 RMeCab RCaBoCha CaBoCha ( )

37 RMeCab RCaBoCha p. 23/29 RMeCab RCaBoCha CaBoCha ( ) Wordnet

38 RMeCab RCaBoCha p. 24/29 footnote

39 RMeCab RCaBoCha p. 25/29 RCaBoCha CaBocha > RCaBoChaFreq(" ") Term Pos Freq

40 RMeCab RCaBoCha p. 25/29 RCaBoCha CaBocha > RCaBoChaFreq(" ") Term Pos Freq > res <- RCaBoChaMx("morikita") > res[res$pos1 %in% c(" + "), 1:5] TERM POS1 POS2 doc1 doc2 doc

41 RMeCab RCaBoCha p. 26/29 footnote RMeCab CaBoCha RCaBoCha R

42 RMeCab RCaBoCha p. 27/29 WordNet (SQLite DB ) > RWordNet( " " ) " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "

43 RMeCab RCaBoCha p. 28/29 footnote Wordnet R RCaBoCha

44 RMeCab RCaBoCha p. 29/29 GUI Rcommander

() (MeCab) 1 Juman ChaSen 2 MeCab ChaSen 1.3 MeCab MeCab OS Windows MeCab [] [Binary package for MS-Windows] [] sourceforge.net [mecab-win32] Mac OS

() (MeCab) *1 Juman ChaSen *2 MeCab ChaSen 1.3 MeCab MeCab OS Windows MeCab [] [Binary package for MS-Windows] [] sourceforge.net [mecab-win32] Mac OS RMeCab 2008 6 16 1 MeCab RMeCab 1 1.1.............................................. 1 1.2............................................ 1 1.3 MeCab......................................... 2 1.4 RMeCab..........................................