数据框字符串拆分为列而不是行[重复]

Question

这个问题在这里已有答案：

• Split delimited strings in a column and insert as new rows [duplicate] 6个答案

我有这种格式的数据框：

df <- data.frame(names= c('perform data cleansing','information categorisation'))

                      names
1     perform data cleansing
2 information categorisation

我想获得这种格式：

                      names             tokens
1      perform data cleansing           perform
1      perform data cleansing              data
1      perform data cleansing         cleansing
2  information categorisation       information
2  information categorisation    categorisation

Answer 1

我喜欢tidyr::unnest：

library(dplyr)
library(tidyr)
df %>% mutate(tokens = strsplit(as.character(names), split = " ")) %>%
  unnest()
# names         tokens
# 1     perform data cleansing        perform
# 2     perform data cleansing           data
# 3     perform data cleansing      cleansing
# 4 information categorisation    information
# 5 information categorisation categorisation

但你也可以在base做到这一切：

tokens = strsplit(as.character(df$names), split = " ")
result = data.frame(names = rep(df$names, lengths(tokens)),
                    tokens = unlist(tokens),
                    stringsAsFactors = FALSE)
#                        names         tokens
# 1     perform data cleansing        perform
# 2     perform data cleansing           data
# 3     perform data cleansing      cleansing
# 4 information categorisation    information
# 5 information categorisation categorisation

一个带有额外功能进行文本分析的版本是tidytext::unnest_tokens：

df$names = as.character(df$names)
tidytext::unnest_tokens(df, output = tokens, input = names, drop = FALSE)
#                          names         tokens
# 1       perform data cleansing        perform
# 1.1     perform data cleansing           data
# 1.2     perform data cleansing      cleansing
# 2   information categorisation    information
# 2.1 information categorisation categorisation