數據框r和二進制列的一列中使用的最高單詞

Question

我有一個這樣的數據框

df=  data.frame(
    text= c("test and run", "rest and sleep", "test", "test of course"), 
    id = c('a','b','c','d'))
#            text id
#1   test and run  a
#2 rest and sleep  b
#3           test  c
#4 test of course  d

我想要

1. 使用緊湊的方式（無循環）來獲取列文本中前2個重復次數最高的單詞（“測試” 3-“和” 2）

2. 創建/添加與前2個值匹配的二進制列。

topTextBinary 
1, 1
0, 1
1, 0
1, 0

用於“測試”，“和”

            text id topTextBinary
1   test and run  a          1, 1
2 rest and sleep  b          0, 1
3           test  c          1, 0
4 test of course  d          1, 0

謝謝

---

R Studio版本

platform       x86_64-w64-mingw32          
arch           x86_64                      
os             mingw32                     
system         x86_64, mingw32             
status                                     
major          3                           
minor          4.3                         
year           2017                        
month          11                          
day            30                          
svn rev        73796                       
language       R                           
version.string R version 3.4.3 (2017-11-30)
nickname       Kite-Eating Tree      

Answer 1

我們可以執行以下操作：

# Word frequency table
tbl <- table(unlist(strsplit(as.character(df$text), " ")));

# Top 2 words
top <- tbl[order(tbl, decreasing = T)][1:2];

# Flag top2 words per row
library(tidyverse);
map(names(top), ~ df %>%
    mutate(!!.x := as.numeric(grepl(.x, text)))) %>%
    reduce(left_join)
#Joining, by = c("text", "id")
#            text id test and
#1   test and run  a    1   1
#2 rest and sleep  b    0   1
#3           test  c    1   0
#4 test of course  d    1   0

或unite從2分二進制列項目分成單個列：

map(names(top), ~ df %>%
    mutate(!!.x := as.numeric(grepl(.x, text)))) %>%
    reduce(left_join) %>%
    unite(topTextBinary, -(1:2), sep = ", ");
#            text id topTextBinary
#1   test and run  a          1, 1
#2 rest and sleep  b          0, 1
#3           test  c          1, 0
#4 test of course  d          1, 0

Answer 2

使用Base R：

top2=names(sort(table(unlist(strsplit(as.character(df$text),"\\s"))),T))[1:2]

transform(df,m=paste(grepl(top2[1],text)+0,grepl(top2[2],text)+0,sep=","))
            text id   m
1   test and run  a 1,1
2 rest and sleep  b 0,1
3           test  c 1,0
4 test of course  d 1,0

如果目標是使用3,4，甚至前10個單詞，那么您可以考慮執行以下操作：

transform(df,m=do.call(paste,c(sep=",",data.frame(t(outer(top2,df$text,Vectorize(grepl))+0L)))))
            text id   m
1   test and run  a 1,1
2 rest and sleep  b 0,1
3           test  c 1,0
4 test of course  d 1,0