使用dplyr选择并绘制最高频率 [英] Select and plot top frequencies with dplyr

问题描述

目标是选择/过滤数据框中频率最高(发生次数最多)的前3个(或n个)事件，然后使用ggplot2中的小节图绘制这些事件.

The objective is to select/filter top 3 (or n) events that have the largest frequencies (occurrences) in a dataframe then plot these using a barplot in ggplot2.

示例:

library(dplyr)
df <- data.frame(
  type=c("car","bike","horse","boat","yacht","train"),freq=c(20,2,5,60,11,10))

到目前为止，我可以安排 df :

So far, I could arrange df:

df_order <- df %>% 
            arrange(desc(freq))
[1] df_order
type freq
1  boat   60
2   car   20
3 yacht   11
4 train   10
5 horse    5
6  bike    2 

理想的结果是仅选择前3种类型，然后使用小节图绘制这些类型.我认为 count 将很有用，但不确定如何做到这一点.有什么想法吗?

The desired result is to select only the top 3 types then plot these using a barplot. I think count will be useful, but not sure how to do that. Any ideas?

推荐答案

根据"freq"列( arrange(...))对数据集进行排序后，我们可以排在前3位使用 slice 的值，使用 ggplot ，在 aes 中指定'x'和'y'变量，并使用geom_bar

After we order the dataset based on the 'freq' column (arrange(...)), we can the top 3 values with slice, use ggplot, specify the 'x' and 'y' variables in the aes, and plot the bar with geom_bar

 library(ggplot2)
 library(dplyr)
 df %>% 
    arrange(desc(freq)) %>%
    slice(1:3) %>%
    ggplot(., aes(x=type, y=freq))+
              geom_bar(stat='identity')

或者另一个选择是 top_n ，这是一个方便的包装器，它使用 filter 和 min_rank 选择顶部的'n'(3)个观测值在频率"列中，然后如上所述使用 ggplot .

Or another option is top_n which is a convenient wrapper that uses filter and min_rank to select the top 'n' (3) observations in 'freq' column and use ggplot as above.

top_n(df, n=3, freq) %>%
          ggplot(., aes(x=type, y=freq))+
              geom_bar(stat='identity')

这篇关于使用dplyr选择并绘制最高频率的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！