如何计算 R 中 dataframe 中一组其他唯一值的最小/最大项？

Question

I have data set in R from Billboard top hits. I am able to count the number of unique hits for a given artist (see code below) but having trouble figuring out how to find the highest point on the charts a song went. The only thing I can think of is that before I filter out the unique values to run a loop for each song and calculate the min value. I am new to R so unaware of other easier ways.

mydata=read.csv("Hot100.csv")
mydata <- mydata[order(mydata$artist, mydata$song, mydata$date),]
head(mydata)

# date position        song  artist
# 218482 2000-07-01       40 Bye Bye Bye 'N Sync
# 226912 2002-02-09       70  Girlfriend 'N Sync
# 226997 2002-02-16       55  Girlfriend 'N Sync
# 227072 2002-02-23       30  Girlfriend 'N Sync
# 227164 2002-03-02       22  Girlfriend 'N Sync
# 227260 2002-03-09       18  Girlfriend 'N Sync

# to remove some cols - leaves artist and song.  Has duplicates
mysub = subset(mydata, select = -c(date, position))

# now to make unique
mysub_u = unique(mysub[,c(1,2)])
View(mysub_u)

# put into table form
mytable = table(mysub_u$artist)

# but this is table form , not df
df=as.data.frame(mytable)

head(df)

# Var1 Freq
# 1                  'N Sync    7
# 2 'N Sync & Gloria Estefan    1
# 3  'N Sync Featuring Nelly    1
# 4             'Til Tuesday    1
# 5      "Weird Al" Yankovic    2
# 6                    (+44)    1

How could I great a table which would list the artist, song and highest number (position) it went to, with 1 being the highest?

Answer 1

It would have been nice to have a bigger dataset (or given any usable data) to play with. However, here is a way to do demonstrated on the small data you provided.

library(readr)
library(dplyr)

mydata <- read_table2("index date position song artist
218482 2000-07-01 40 Bye_Bye_Bye 'N_Sync
226912 2002-02-09 70  Girlfriend 'N_Sync
226997 2002-02-16 55  Girlfriend 'N_Sync
227072 2002-02-23 30  Girlfriend 'N_Sync
227164 2002-03-02 22  Girlfriend 'N_Sync
227260 2002-03-09 18  Girlfriend 'N_Sync")

out <- mydata %>% 
  group_by(artist,song) %>% 
  mutate(highest_position = min(position)) %>% 
  select(-index,-date,-position) %>% 
  unique(.)

Output:

> out
# A tibble: 2 x 3
# Groups:   artist, song [2]
  song        artist  highest_position
  <chr>       <chr>              <dbl>
1 Bye_Bye_Bye 'N_Sync               40
2 Girlfriend  'N_Sync               18