*disclaimer
603615
quanteda: Quantitative Analysis of Textual Data
注意 quanteda v3参照のこと
https://rdrr.io/cran/quanteda/
install.packages("quanteda", dependencies=T) library(quanteda)
コーパスを構築してコーパスデータとして扱う
- tmpにテキストファイルが入っているとして
tmp.cor <- corpus(tmp)
コーパスデータの概要を見る
summary(tmp.cor)
コーパスデータの中身を見る
texts(tmp.cor)
KWIC検索をする kwic()
kwick(tmp.cor, pattern = "文字列")
トークンのリスト作成
tokens(tmp)
- オプション
remove_numbers=T
- オプション
remove_punct=T
- 文単位
what="sentence"
- 単語単位
what="word"
- 文字単位
what="char"
> tmp.cor <- corpus(tmp) > tmp.cor Corpus consisting of 1 document and 0 docvars. > summary(tmp.cor) Corpus consisting of 1 document: Text Types Tokens Sentences text1 120 233 9 Source: C:/Users/ /Documents/* on x86-64 by sugiura Created: Thu Nov 14 14:40:11 2019 Notes: > texts(tmp.cor) text1 "Computer Terminal Systems Inc said\nit has completed the sale of 200,000 shares of its (中略) generated labels, forms,\ntags and ticket printers and terminals.\n Reuter" > kwic(tmp.cor, pattern="for") [text1, 39] > of Lugano, Switzerland | for | 50,000 dlrs. The company [text1, 50] said the warrants are exercisable | for | five years at a purchase [text1, 171] of Houston, Tex. | for | 200,000 dlrs. But, [text1, 191] worldwide licensee of the technology | for | Woodco. The company said > tokens(tmp) tokens from 1 document. text1 : [1] "Computer" "Terminal" "Systems" "Inc" "said" "it" [7] "has" "completed" "the" "sale" "of" "200,000" [13] "shares" "of" "its" "common" "stock" "," [19] "and" "warrants" "to" "acquire" "an" "additional" > tokens(tmp, remove_numbers=T, remove_punct=T) tokens from 1 document. text1 : [1] "Computer" "Terminal" "Systems" "Inc" "said" "it" [7] "has" "completed" "the" "sale" "of" "shares" [13] "of" "its" "common" "stock" "and" "warrants" [19] "to" "acquire" "an" "additional" "one" "mln" [25] "shares" "to" "Sedio" "N.V" "of" "Lugano" > tokens(tmp, what="sentence") tokens from 1 document. text1 : [1] "Computer Terminal Systems Inc said it has completed the sale of 200,000 shares of its common stock, and warrants to acquire an additional one mln shares, to <Sedio N.V.> of Lugano, Switzerland for 50,000 dlrs." [2] "The company said the warrants are exercisable for five years at a purchase price of .125 dlrs per share." [3] "Computer Terminal said Sedio also has the right to buy additional shares and increase its total holdings up to
quanteda.corpora
https://github.com/quanteda/quanteda.corpora
- reference
https://data.library.virginia.edu/a-beginners-guide-to-text-analysis-with-quanteda/
https://sugiura-ken.org/wiki/