r - 将单词(来自定义列表)分组为 R 中的主题

Question

我是 Stackoverflow 的新手，正在尝试学习 R。

我想在文本中找到一组已定义的词。以我定义的相关主题的表格格式返回这些单词的计数。

这是我的尝试:

text <- c("Green fruits are such as apples, green mangoes and avocados are good for high blood pressure. Vegetables range from greens like lettuce, spinach, Swiss chard, and mustard greens are great for heart disease. When researchers combined findings with several other long-term studies and looked at coronary heart disease and stroke separately, they found a similar protective effect for both. Green mangoes are the best.")

library(qdap)

**#Own Defined Lists**

fruit <- c("apples", "green mangoes", "avocados") 
veg <- c("lettuce", "spinach", "Swiss chard", "mustard greens")

**#Splitting in Sentences**

stext <- strsplit(text, split="\\.")[[1]] 

**#Obtain and Count Occurences**
library(plyr) 
fruitres <- laply(fruit, function(x) grep(x, stext))
vegres <- laply(veg, function(x) grep(x, stext))

**#Quick check, and not returning 2 results for** "green mangoes"
grep("green mangoes", stext)

**#Trying with stringr package**
tag_ex <- paste0('(', paste(fruit, collapse = '|'), ')')
tag_ex

library(dplyr)
library(stringr)


themes = sapply(str_extract_all(stext, tag_ex), function(x) paste(x, collapse=','))[[1]]
themes     


#Create data table
library(data.table)
data.table(fruit,fruitres)

使用各自的 qdap 和 stringr 包我无法获得我想要的解决方案。

将水果和蔬菜组合在一张 table 上的理想解决方案

apples               fruit     1
green mangoes        fruit     2
avocados             fruit     1
lettuce              veg       1
spinach              veg       1
Swiss chard          veg       1
mustard greens       veg       1

任何帮助将不胜感激。谢谢

Answer 1

我尝试对 N 个向量进行泛化

tidyverse 和 stringr 解决方案

library(tidyverse)
library(stringr)

为你的向量创建一个data.frame

data <- c("fruit","veg")   # vector names
L <- map(data, ~get(.x))
names(L) <- data
long <- map_df(1:length(L), ~data.frame(category=rep(names(L)[.x]), type=L[[.x]]))

# You may receive warnings about coercing to characters

#   category           type
# 1    fruit         apples
# 2    fruit  green mangoes
# 3    fruit       avocados
# etc

计算每个的实例

long %>%
  mutate(count=str_count(tolower(text), tolower(type)))

输出

  category           type count
1    fruit         apples     1
2    fruit  green mangoes     2
3    fruit       avocados     1
4      veg        lettuce     1
# etc

额外内容

我们可以很容易地添加另一个向量

health <- c("blood", "heart")
data <- c("fruit","veg", "health")

# code as above

额外输出(tail)

6      veg    Swiss chard     1
7      veg mustard greens     1
8   health          blood     1
9   health          heart     2