Introduction

Welcome to my rmd. In this Learning By Building task, I would like to do data visualization of USvideos.csv

Data Explanatory

Data Input

options(scipen = 0)
options(warn = -1)
vids <- read.csv("USvideos.csv", stringsAsFactors = FALSE)
vids <- vids[match(unique(vids$title),vids$title),]
vids

USvideos.csv contains of column:

trending_date: tanggal trending
title: judul video
channel_title: nama channel Youtube
category_id: kategori video
publish_time: tanggal upload video
views: jumlah views
likes: jumlah likes
dislikes: jumlah dislikes
comment_count: jumlah komentar
comment_disabled: apakah kolom komentar tidak diaktifkan
rating_disabled: apakah rating video tidak diaktifkan
video_error_or_removed: apakah video dihapus

Check Missing Data

No Missing Value in data USvideos.csv

colSums(is.na(vids))

##          trending_date                  title          channel_title 
##                      0                      0                      0 
##            category_id           publish_time                  views 
##                      0                      0                      0 
##                  likes               dislikes          comment_count 
##                      0                      0                      0 
##      comments_disabled       ratings_disabled video_error_or_removed 
##                      0                      0                      0

Data Type Adjustment

Change data type of column title, and channel title to factor using lapply

vids[,c("title", "channel_title")] <- lapply(vids[,c("title", "channel_title")], as.factor)
str(vids)

## 'data.frame':    2986 obs. of  12 variables:
##  $ trending_date         : chr  "17.14.11" "17.14.11" "17.14.11" "17.14.11" ...
##  $ title                 : Factor w/ 2986 levels "'I have dad moves': Barack Obama discusses dancing on David Letterman's new Netflix show",..: 2802 2574 2081 1903 1231 89 2164 143 2482 2920 ...
##  $ channel_title         : Factor w/ 1404 levels "_¢_Á_\235","“÷\201\220µ_‘â¬“_\220 Korean Englishman",..: 193 682 1042 469 898 556 1059 280 6 1354 ...
##  $ category_id           : int  22 24 23 24 24 28 24 28 1 25 ...
##  $ publish_time          : chr  "2017-11-13T17:13:01.000Z" "2017-11-13T07:30:00.000Z" "2017-11-12T19:05:24.000Z" "2017-11-13T11:00:04.000Z" ...
##  $ views                 : int  748374 2418783 3191434 343168 2095731 119180 2103417 817732 826059 256426 ...
##  $ likes                 : int  57527 97185 146033 10172 132235 9763 15993 23663 3543 12654 ...
##  $ dislikes              : int  2966 6146 5339 666 1989 511 2445 778 119 1363 ...
##  $ comment_count         : int  15954 12703 8181 2146 17518 1434 1970 3432 340 2368 ...
##  $ comments_disabled     : logi  FALSE FALSE FALSE FALSE FALSE FALSE ...
##  $ ratings_disabled      : logi  FALSE FALSE FALSE FALSE FALSE FALSE ...
##  $ video_error_or_removed: logi  FALSE FALSE FALSE FALSE FALSE FALSE ...

Change data type of column trending_date, publish_time to date using lubridate

library(lubridate)
vids$trending_date <- ydm(vids$trending_date)
vids$publish_time <- ymd_hms(vids$publish_time, tz = "America/New_York")
head(vids)

Change code of category_id

vids$category_id <- sapply(as.character(vids$category_id), switch, 
                           "1" = "Film and Animation",
                           "2" = "Autos and Vehicles", 
                           "10" = "Music", 
                           "15" = "Pets and Animals", 
                           "17" = "Sports",
                           "19" = "Travel and Events", 
                           "20" = "Gaming", 
                           "22" = "People and Blogs", 
                           "23" = "Comedy",
                           "24" = "Entertainment", 
                           "25" = "News and Politics",
                           "26" = "Howto and Style", 
                           "27" = "Education",
                           "28" = "Science and Technology", 
                           "29" = "Nonprofit and Activism",
                           "43" = "Shows")

vids$category_id <- as.factor(vids$category_id)
head(vids,5)

Data Analysis

Most View Movie Title

most_view_title <- vids %>% 
  group_by(title) %>%
  summarise(Views =  sum(views))
most_view_title <- head(most_view_title[order(most_view_title$Views, decreasing = T),],5)
most_view_title <- most_view_title %>% 
  mutate(remark=paste(title,":",Views))

most_view_title_plot<- most_view_title %>% 
      ggplot(aes(x=Views,y=reorder(title,Views),text=remark)) +
      geom_col(aes(fill = title),show.legend = F) +
      labs(title = "Most View Title",x="Views",y="Title") +
      theme_algoritma+theme(legend.position = "none")
    
ggplotly(most_view_title_plot,tooltip = "text")

2. Most View Channel_Title

most_view_channel_title <- vids %>% 
  group_by(channel_title) %>%
  summarise(Views =  sum(views))
most_view_channel_title <- head(most_view_channel_title[order(most_view_channel_title$Views, decreasing = T),],5)
most_view_channel_title <- most_view_channel_title %>% 
  mutate(remark=paste(channel_title,":",Views))

most_view_channel_title_plot<- most_view_channel_title %>% 
      ggplot(aes(x=Views,y=reorder(channel_title,Views),text=remark)) +
      geom_col(aes(fill = channel_title),show.legend = F) +
      labs(title = "Most View Channel Title",x="Views",y="Channel Title") +
      theme_algoritma+theme(legend.position = "none")
    
ggplotly(most_view_channel_title_plot,tooltip = "text")

3. Most View Category

most_view_categoryid <- vids %>% 
  group_by(category_id) %>%
  summarise(Views =  sum(views))
most_view_categoryid <- head(most_view_categoryid[order(most_view_categoryid$Views, decreasing = T),],5)
most_view_categoryid <- most_view_categoryid %>% 
  mutate(remark=paste(category_id,":",Views))

most_view_categoryid_plot<- most_view_categoryid %>% 
      ggplot(aes(x=Views,y=reorder(category_id,Views),text=remark)) +
      geom_col(aes(fill = category_id),show.legend = F) +
      labs(title = "Most View Category",x="Views",y="Category") +
      theme_algoritma+theme(legend.position = "none")
    
ggplotly(most_view_categoryid_plot,tooltip = "text")

4. Most Likes Movie Title

most_likes_title <- vids %>% 
  group_by(title) %>%
  summarise(Likes =  sum(likes))
most_likes_title <- head(most_likes_title[order(most_likes_title$Likes, decreasing = T),],5)
most_likes_title <- most_likes_title %>% 
  mutate(remark=paste(title,":",Likes))

most_likes_title_plot<- most_likes_title %>% 
      ggplot(aes(x=Likes,y=reorder(title,Likes),text=remark)) +
      geom_col(aes(fill = title),show.legend = F) +
      labs(title = "Most Likes Title",x="Likes",y="Title") +
      theme_algoritma+theme(legend.position = "none")
    
ggplotly(most_likes_title_plot,tooltip = "text")

5. Most Likes Channel_Title

most_likes_channel_title <- vids %>% 
  group_by(channel_title) %>%
  summarise(Likes =  sum(likes))
most_likes_channel_title <- head(most_likes_channel_title[order(most_likes_channel_title$Likes, decreasing = T),],5)
most_likes_channel_title <- most_likes_channel_title %>% 
  mutate(remark=paste(channel_title,":",Likes))

most_likes_channel_title_plot<- most_likes_channel_title %>% 
      ggplot(aes(x=Likes,y=reorder(channel_title,Likes),text=remark)) +
      geom_col(aes(fill = channel_title),show.legend = F) +
      labs(title = "Most Likes Channel Title",x="Likes",y="Channel Title") +
      theme_algoritma+theme(legend.position = "none")
    
ggplotly(most_likes_channel_title_plot,tooltip = "text")

6. Most Likes Category

most_likes_categoryid <- vids %>% 
  group_by(category_id) %>%
  summarise(Likes =  sum(likes))
most_likes_categoryid <- head(most_likes_categoryid[order(most_likes_categoryid$Likes, decreasing = T),],5)
most_likes_categoryid <- most_likes_categoryid %>% 
  mutate(remark=paste(category_id,":",Likes))

most_likes_categoryid_plot<- most_likes_categoryid %>% 
      ggplot(aes(x=Likes,y=reorder(category_id,Likes),text=remark)) +
      geom_col(aes(fill = category_id),show.legend = F) +
      labs(title = "Most Likes Category",x="LIkes",y="Category") +
      theme_algoritma+theme(legend.position = "none")
    
ggplotly(most_likes_categoryid_plot,tooltip = "text")

7. Most Dislikes Movie Title

most_dislikes_title <- vids %>% 
  group_by(title) %>%
  summarise(Dislikes =  sum(dislikes))
most_dislikes_title <- head(most_dislikes_title[order(most_dislikes_title$Dislikes, decreasing = T),],5)
most_dislikes_title <- most_dislikes_title %>% 
  mutate(remark=paste(title,":",Dislikes))

most_dislikes_title_plot<- most_dislikes_title %>% 
      ggplot(aes(x=Dislikes,y=reorder(title,Dislikes),text=remark)) +
      geom_col(aes(fill = title),show.legend = F) +
      labs(title = "Most Dislikes Title",x="Dislikes",y="Title") +
      theme_algoritma+theme(legend.position = "none")
    
ggplotly(most_dislikes_title_plot,tooltip = "text")

8. Most Dislikes Channel_Title

most_dislikes_channel_title <- vids %>% 
  group_by(channel_title) %>%
  summarise(Dislikes =  sum(dislikes))
most_dislikes_channel_title <- head(most_dislikes_channel_title[order(most_dislikes_channel_title$Dislikes, decreasing = T),],5)
most_dislikes_channel_title <- most_dislikes_channel_title %>% 
  mutate(remark=paste(channel_title,":",Dislikes))

most_dislikes_channel_title_plot<- most_dislikes_channel_title %>% 
      ggplot(aes(x=Dislikes,y=reorder(channel_title,Dislikes),text=remark)) +
      geom_col(aes(fill = channel_title),show.legend = F) +
      labs(title = "Most Dislikes Channel Title",x="Dislikes",y="Channel Title") +
      theme_algoritma+theme(legend.position = "none")
    
ggplotly(most_dislikes_channel_title_plot,tooltip = "text")

9. Most Dislikes Category

most_dislikes_categoryid <- vids %>% 
  group_by(category_id) %>%
  summarise(Dislikes =  sum(dislikes))
most_dislikes_categoryid <- head(most_dislikes_categoryid[order(most_dislikes_categoryid$Dislikes, decreasing = T),],5)
most_dislikes_categoryid <- most_dislikes_categoryid %>% 
  mutate(remark=paste(category_id,":",Dislikes))

most_dislikes_categoryid_plot<- most_dislikes_categoryid %>% 
      ggplot(aes(x=Dislikes,y=reorder(category_id,Dislikes),text=remark)) +
      geom_col(aes(fill = category_id),show.legend = F) +
      labs(title = "Most Dislikes Category",x="Dislikes",y="Category") +
      theme_algoritma+theme(legend.position = "none")
    
ggplotly(most_dislikes_categoryid_plot,tooltip = "text")

10. Most Commented Movie Title

most_commented_title <- vids %>% 
  group_by(title) %>%
  summarise(Commented =  sum(comment_count))
most_commented_title <- head(most_commented_title[order(most_commented_title$Commented, decreasing = T),],5)
most_commented_title <- most_commented_title %>% 
  mutate(remark=paste(title,":",Commented))

most_commented_title_plot<- most_commented_title %>% 
      ggplot(aes(x=Commented,y=reorder(title,Commented),text=remark)) +
      geom_col(aes(fill = title),show.legend = F) +
      labs(title = "Most Commented Title",x="Total Comment",y="Title") +
      theme_algoritma+theme(legend.position = "none")
    
ggplotly(most_commented_title_plot,tooltip = "text")

11. Most Commented Channel_Title

most_commented_channel_title <- vids %>% 
  group_by(channel_title) %>%
  summarise(Commented =  sum(comment_count))
most_commented_channel_title <- head(most_commented_channel_title[order(most_commented_channel_title$Commented, decreasing = T),],5)
most_commented_channel_title <- most_commented_channel_title %>% 
  mutate(remark=paste(channel_title,":",Commented))

most_commented_channel_title_plot<- most_commented_channel_title %>% 
      ggplot(aes(x=Commented,y=reorder(channel_title,Commented),text=remark)) +
      geom_col(aes(fill = channel_title),show.legend = F) +
      labs(title = "Most Commented Channel Title",x="Total Comment",y="Channel Title") +
      theme_algoritma+theme(legend.position = "none")
    
ggplotly(most_commented_channel_title_plot,tooltip = "text")

Most Commented Category

most_commented_categoryid <- vids %>% 
  group_by(category_id) %>%
  summarise(Commented =  sum(comment_count))
most_commented_categoryid <- head(most_commented_categoryid[order(most_commented_categoryid$Commented, decreasing = T),],5)
most_commented_categoryid <- most_commented_categoryid %>% 
  mutate(remark=paste(category_id,":",Commented))

most_commented_categoryid_plot<- most_commented_categoryid %>% 
      ggplot(aes(x=Commented,y=reorder(category_id,Commented),text=remark)) +
      geom_col(aes(fill = Commented),show.legend = F) +
      labs(title = "Most Commented Category",x="Total Comment",y="Category") +
      theme_algoritma+theme(legend.position = "none")
    
ggplotly(most_commented_categoryid_plot,tooltip = "text")

11. Average Duration from Publish to Trending Per Category

vids$publish_date <- date(vids$publish_time)
vids$timetotrend <- vids$trending_date-vids$publish_date
vids$timetotrend2 <- as.factor(ifelse(vids$timetotrend <= 7, vids$timetotrend, "8+"))
avg_duration <- vids %>% 
  group_by(category_id) %>%
  summarise(Duration =  mean(timetotrend))
avg_duration <- avg_duration %>% 
  mutate(remark=paste(category_id,":",Duration))

avg_duration_plot<- avg_duration %>% 
      ggplot(aes(x=Duration,y=reorder(category_id,-Duration),text=remark)) +
      geom_col(aes(fill = Duration),show.legend = F) +
      labs(title = "Average Duration Publish to Trending Per Category",x="Category",y="Duration") +
      theme_algoritma+theme(legend.position = "none")
    
ggplotly(avg_duration_plot,tooltip = "text")

12. Top 15 Trending Channel in US based on Video Count

# buat data frame (aggregasi)
vids.chan <- data.frame(table(vids$channel_title))
vids.chan <- vids.chan[vids.chan$Freq >= 10,]
vids.chan <- vids.chan[order(vids.chan$Freq, decreasing = T),]
vids.chan$Var1 <- droplevels(vids.chan$Var1) 
vids.chan$remark <- paste(vids.chan$Var1,":",vids.chan$Freq)

# menambah keterangan (labs) + percantik plot

p3 <- ggplot(data = vids.chan[1:10,], mapping = aes(x = reorder(Var1, Freq), y = Freq,text=remark)) + 
  geom_col(mapping = aes(fill = Freq)) + 
  coord_flip() + 
  geom_label(mapping = aes(label = Freq), 
            data = vids.chan[1:3,], 
            nudge_y = 1, 
            size = 4) +
  scale_fill_gradient(low = "skyblue",
                      high = "navy") +
  labs(title = "Top 10 Trending Channel in US",
       subtitle = "Based on Video Count",
       x = "",
       y = "Video Count") +
  theme_minimal() + # menambah theme bawaan
  theme(legend.position = "none", # menghilaggkan legend
        panel.grid = element_blank(), # menghilangkan grid
        plot.title = element_text(face = "bold.italic")) + # mengatur font plot.title
  geom_hline(yintercept = 20) # menambah horizontal line (axis y)
ggplotly(p3,tooltip = "text")

Top 10 channel that have video count more than 20
Channel that has the most video counts is Refinery29 with video count 31

13 Relation between Likes vs View in Music Category

library(ggthemes)
vids$commentratio <- vids$comment_count/vids$views
vids$likeratio <- vids$likes/vids$views
vids.music <- vids[vids$category_id == "Music",]

ggplot(vids.music, aes(x = likeratio, y = commentratio)) +
  geom_point(aes(col = timetotrend2, size = dislikes)) +
  labs(title = "",
       subtitle = "",
       x = "likes per view",
       y = "comment per view") +
  guides(size = "none") +
  scale_color_viridis_d() +
  theme_classic() +
  theme(legend.position = "top")

14. We want to visualize proportion total video that was published in certain time (publish_when) in each category.

# data wrangling
vids$publish_hour <- hour(vids$publish_time)

pw <- function(x){
    if(x < 8){
      x <- "12am to 8am"
    }else if(x >= 8 & x < 16){
      x <- "8am to 4pm"
    }else{
      x <- "4pm to 12am"
    }  
}

vids$publish_when <- as.factor(sapply(vids$publish_hour, pw))
vids.mul2 <- as.data.frame(table(category = vids$category_id, # menamakan kolom pada table
                                 publish_time = vids$publish_when))

ggplot(data = vids.mul2, mapping = aes(x = reorder(category, Freq), y = Freq)) +
  geom_col(mapping = aes(fill = publish_time), position = "fill") +
  coord_flip() +
  labs(x = "",
       y = "Video Count",
       fill = "",
       title = "Proportion of YouTube Trending Videos",
       subtitle = "Categories vs. Publish Hour") +
  scale_fill_brewer(palette = "Set2") +
  theme_minimal() +
  theme(legend.position = "top")

Insight:

Shows only publish videos at 4pm-12am
Generally the most trending videos is published at 8am-4pm

Viewers activity on trending videos (based on most views)

vids.trend <- vids[vids$category_id %in% c("Entertainment", "Music", "Comedy"),]
vids.tagg <- aggregate(views ~ publish_hour + category_id, data = vids.trend, FUN = mean)
# visualisasi geom_line
ggplot(data = vids.tagg, 
       mapping = aes(x = publish_hour,
                     y = views)) + # grouping line berdasarkan
  geom_line(aes(col = category_id), lwd = 0.9) +
  facet_wrap(facets = ~category_id, nrow = 3) +
  scale_x_continuous(breaks = seq(from = 0, to = 23, by = 3)) +
  scale_y_continuous(labels = scales::comma) +
  theme_minimal()

Conclusion

Most view

Based on movie title = Marvel Studios Avengers: Infinity War Official Trailer.Total views 37736281.
Based on channel Title = Ed Sheeran. Total views 47640195.
Based on category = Entertainment. Total views 445350694.

Most likes

Based on movie title = BTS MIC Drop (Steve Aoki Remix) official MV. Total likes 2055137.
Based on channel title = ibighit. Total likes 2844877.
Based on category = Music. Total likes 28024864.

Most dislikes

Based on movie title = So Sorry. Total dislikes 629120.
Based on Channel Title = Logan Paul Vlogs. Total dislikes 715595.
Based on category = Entertainment. Total dislikes 1695578.

Most commented

Based on title = So Sorry. Total comment 733373
Based on channel title = Logan Paul Vlogs. Total Comment 868532
Based on category = Entertainment. Total Comment 3315663

Category Shows has the fastest average duration from publish to trending date, While category Film and Animation has the slowest time.
Top 10 channel that have video count more than 20 are Refinery29, The Tonight SHow Starrig Jimmy Fallon, Vox, TheEllenShow, Netflix, NFL, Jimmy Kimmel Live, ESPN, The Late Show with Stephen Colbert, Late Night with Seth Meyers.
Channel that has the most video counts is Refinery29 with video count 31
There is a correlation or relation between likes and views in music category. The highest like the highest comment will happened in category music.
Shows category only publish videos at 4pm-12am
Generally the most trending videos is published at 8am-4pm
Most trending category based on most views are COmedy, Entertainment, and Music. Viewers in category comedy is most active at around 3 o’clock. Viewers in category entertainment is most active at around 8 and 20 o’clock. Viewers in category Music is most active at 21 o’clock

LBB DV - USvideos

Debrina

1/30/2021