LBB 2 - USVID
mochiyam sjarif
3/21/2022
LBB DATA VISUALIZATION
1. Intro
1.1 Greetings
Hi !! Welcome to my Rmd :) in this LBB i will use previous data which is USvideos.csv
1.2 Data’s Brief
This data is about xxxxxxx for listing in USA. and the first thing i need to do is load all package tht might be needed for this dataset.
2. Data Explanatory
2.1 Data Input & Structure
yt <- read.csv("youtubetrends.csv")Futhermore we will inspect our data
dim(yt)## [1] 2986 16head(yt)## trending_date title
## 1 2017-11-14 WE WANT TO TALK ABOUT OUR MARRIAGE
## 2 2017-11-14 The Trump Presidency: Last Week Tonight with John Oliver (HBO)
## 3 2017-11-14 Racist Superman | Rudy Mancuso, King Bach & Lele Pons
## 4 2017-11-14 Nickelback Lyrics: Real or Fake?
## 5 2017-11-14 I Dare You: GOING BALD!?
## 6 2017-11-14 2 Weeks with iPhone X
## channel_title category_id publish_time views
## 1 CaseyNeistat People and Blogs 2017-11-13 12:13:01 748374
## 2 LastWeekTonight Entertainment 2017-11-13 02:30:00 2418783
## 3 Rudy Mancuso Comedy 2017-11-12 14:05:24 3191434
## 4 Good Mythical Morning Entertainment 2017-11-13 06:00:04 343168
## 5 nigahiga Entertainment 2017-11-12 13:01:41 2095731
## 6 iJustine Science and Technology 2017-11-13 14:07:23 119180
## likes dislikes comment_count comments_disabled ratings_disabled
## 1 57527 2966 15954 FALSE FALSE
## 2 97185 6146 12703 FALSE FALSE
## 3 146033 5339 8181 FALSE FALSE
## 4 10172 666 2146 FALSE FALSE
## 5 132235 1989 17518 FALSE FALSE
## 6 9763 511 1434 FALSE FALSE
## video_error_or_removed publish_hour publish_when publish_wday timetotrend
## 1 FALSE 12 8am to 3pm Monday 1
## 2 FALSE 2 12am to 8am Monday 1
## 3 FALSE 14 8am to 3pm Sunday 2
## 4 FALSE 6 12am to 8am Monday 1
## 5 FALSE 13 8am to 3pm Sunday 2
## 6 FALSE 14 8am to 3pm Monday 1tail(yt)## trending_date
## 2981 2018-01-21
## 2982 2018-01-21
## 2983 2018-01-21
## 2984 2018-01-21
## 2985 2018-01-21
## 2986 2018-01-21
## title
## 2981 Short-term spending bill fails in the Senate
## 2982 Jessica Chastain Monologue - SNL
## 2983 Legendary Rocker Tom Petty Died From Accidental Overdose Of Medications, Coroner Says
## 2984 Mulan Hair Cut
## 2985 Urban opera singer
## 2986 GET READY WITH ME | Chit Chat with Winnie Harlow
## channel_title category_id publish_time views likes
## 2981 Washington Post News and Politics 2018-01-19 23:43:42 13044 68
## 2982 Saturday Night Live Entertainment 2018-01-21 00:44:30 23484 858
## 2983 CBS Los Angeles News and Politics 2018-01-19 21:10:07 3976 30
## 2984 Aymeric Favard Sports 2014-03-15 14:21:22 16231 103
## 2985 Marcos Horacio Sports 2018-01-19 12:26:11 13531 136
## 2986 LaToya Forever Entertainment 2018-01-18 18:03:24 44754 3246
## dislikes comment_count comments_disabled ratings_disabled
## 2981 22 220 FALSE FALSE
## 2982 80 136 FALSE FALSE
## 2983 5 45 FALSE FALSE
## 2984 7 15 FALSE FALSE
## 2985 5 16 FALSE FALSE
## 2986 28 386 FALSE FALSE
## video_error_or_removed publish_hour publish_when publish_wday timetotrend
## 2981 FALSE 23 3pm to 12am Friday 1
## 2982 FALSE 0 12am to 8am Sunday 0
## 2983 FALSE 21 3pm to 12am Friday 1
## 2984 FALSE 14 8am to 3pm Saturday 8+
## 2985 FALSE 12 8am to 3pm Friday 2
## 2986 FALSE 18 3pm to 12am Thursday 3From our inspection we can conclude :
yt data contain 2986 of rows and 16 of coloumns
Each of column name : “trending_date”, “title”, “channel_title”, “category_id”, “publish_time”, “views”, “dislikes”, “comment_count”, “comments_disabled”, “ratings_disabled”, “video_error_or_removed”, “publish_hour”, “publish_when”, “publish_wday”, & “timetotrend”
str(yt)## 'data.frame': 2986 obs. of 16 variables:
## $ trending_date : chr "2017-11-14" "2017-11-14" "2017-11-14" "2017-11-14" ...
## $ title : chr "WE WANT TO TALK ABOUT OUR MARRIAGE" "The Trump Presidency: Last Week Tonight with John Oliver (HBO)" "Racist Superman | Rudy Mancuso, King Bach & Lele Pons" "Nickelback Lyrics: Real or Fake?" ...
## $ channel_title : chr "CaseyNeistat" "LastWeekTonight" "Rudy Mancuso" "Good Mythical Morning" ...
## $ category_id : chr "People and Blogs" "Entertainment" "Comedy" "Entertainment" ...
## $ publish_time : chr "2017-11-13 12:13:01" "2017-11-13 02:30:00" "2017-11-12 14:05:24" "2017-11-13 06:00:04" ...
## $ 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 ...
## $ publish_hour : int 12 2 14 6 13 14 0 16 9 8 ...
## $ publish_when : chr "8am to 3pm" "12am to 8am" "8am to 3pm" "12am to 8am" ...
## $ publish_wday : chr "Monday" "Monday" "Sunday" "Monday" ...
## $ timetotrend : chr "1" "1" "2" "1" ...as we can see here, we need to adjust some of the data types.
library(lubridate)##
## Attaching package: 'lubridate'## The following objects are masked from 'package:base':
##
## date, intersect, setdiff, unionyt$publish_time <- ymd_hms(yt$publish_time, tz = "America/New_York")yt$trending_date <- date(yt$trending_date)
yt$publish_when <- as.factor(yt$publish_when)
yt$category_id<- as.factor(yt$category_id)
yt$publish_wday <- as.factor(yt$publish_wday)
yt$title <- as.factor(yt$title)
yt$channel_title <- as.factor(yt$channel_title)
yt$timetotrend <- as.factor(yt$timetotrend)2.2 Missing Data
colSums(is.na(yt))## 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
## publish_hour publish_when publish_wday
## 0 0 0
## timetotrend
## 0alright!! there are no missing values!!
2.3 Subseting and Practical Statistics
First, lets check our data’s summary
summary(yt)## trending_date
## Min. :2017-11-14
## 1st Qu.:2017-11-27
## Median :2017-12-14
## Mean :2017-12-15
## 3rd Qu.:2018-01-04
## Max. :2018-01-21
##
## title
## _\xa3_\xbc\x84\xdb_\xa1\x84\xd1__\x84\x81\x84\xce_\xbc_\xa1 _\xa1\x84\xdb__\x84Є\x8f _\xe1__\x84\xd0_Ȅ\xce___\xfc_\xc8_\xa1 ___\xab\x84\xdb_\xa1_\xe1\x84\xc4 _\xab___\xa1 \x84\x81_\xb5_\xc8_\xa1 ___\xa1 _\xc1\x84\xc9___\xab\x84\xd0: 1
## _\xd9_\xc4 How to make Pumpkin Pie Mistakes : 1
## 'I have dad moves': Barack Obama discusses dancing on David Letterman's new Netflix show : 1
## 'I have taken poison' claims war criminal : 1
## 'Lightning Strike' B777-300 on departure @ KL743 // PH-BVS : 1
## 'Lord of the Rings' TV Series Gets Multi-Season Order At Amazon | News Flash | Entertainment Weekly : 1
## (Other) :2980
## channel_title category_id
## Refinery29 : 31 Entertainment :736
## The Tonight Show Starring Jimmy Fallon: 30 Music :391
## Vox : 29 Howto and Style :285
## TheEllenShow : 28 Comedy :273
## Netflix : 27 News and Politics:271
## NFL : 25 People and Blogs :228
## (Other) :2816 (Other) :802
## publish_time views likes
## Min. :2008-04-05 14:22:40 Min. : 687 Min. : 0.0
## 1st Qu.:2017-11-22 10:51:03 1st Qu.: 43014 1st Qu.: 931.2
## Median :2017-12-11 12:41:02 Median : 149196 Median : 4785.0
## Mean :2017-11-07 14:44:26 Mean : 512602 Mean : 25412.1
## 3rd Qu.:2018-01-01 10:22:33 3rd Qu.: 483848 3rd Qu.: 17517.0
## Max. :2018-01-21 00:44:30 Max. :37736281 Max. :2055137.0
##
## dislikes comment_count comments_disabled ratings_disabled
## Min. : 0.0 Min. : 0 Mode :logical Mode :logical
## 1st Qu.: 39.0 1st Qu.: 152 FALSE:2942 FALSE:2976
## Median : 148.5 Median : 589 TRUE :44 TRUE :10
## Mean : 1257.7 Mean : 3545
## 3rd Qu.: 541.0 3rd Qu.: 1964
## Max. :629120.0 Max. :733373
##
## video_error_or_removed publish_hour publish_when publish_wday
## Mode :logical Min. : 0.00 12am to 8am: 552 Friday :478
## FALSE:2985 1st Qu.: 9.00 3pm to 12am: 795 Monday :415
## TRUE :1 Median :12.00 8am to 3pm :1639 Saturday :269
## Mean :11.95 Sunday :247
## 3rd Qu.:16.00 Thursday :490
## Max. :23.00 Tuesday :561
## Wednesday:526
## timetotrend
## 1 :1379
## 2 : 714
## 3 : 290
## 4 : 189
## 5 : 132
## 8+ : 129
## (Other): 153From summary above, we can take a note for some important information:
- The most favourite category is from others category, then Entertainment comes after, the least comes from people and blogs
- Max views from this data has 37736281, while minimum has 687
- Max Likes from this data has 2055137.0, while minimum has 0, while dislike has more max number with 629120.0, and the minimum dislike has the same number as Likes.
- from Publish_when, we can see that from 8am to 3pm has the most number, while from 12 am to 8am has the least.
- Tuesday becomes the favorite day to publish while on Sunday is the least facorite.
3. Study Case
These are the libraries that I am going to use
library(ggplot2)
library(plotly) ##
## Attaching package: 'plotly'## The following object is masked from 'package:ggplot2':
##
## last_plot## The following object is masked from 'package:stats':
##
## filter## The following object is masked from 'package:graphics':
##
## layoutlibrary(glue) - Which category has the most likes and the least likes?
First we have to find the likes ratio, and dislike ratio
yt$like_ratio <-yt$likes/yt$views
yt$dislike_ratio <- yt$dislikes/yt$views
head(yt)## trending_date title
## 1 2017-11-14 WE WANT TO TALK ABOUT OUR MARRIAGE
## 2 2017-11-14 The Trump Presidency: Last Week Tonight with John Oliver (HBO)
## 3 2017-11-14 Racist Superman | Rudy Mancuso, King Bach & Lele Pons
## 4 2017-11-14 Nickelback Lyrics: Real or Fake?
## 5 2017-11-14 I Dare You: GOING BALD!?
## 6 2017-11-14 2 Weeks with iPhone X
## channel_title category_id publish_time views
## 1 CaseyNeistat People and Blogs 2017-11-13 12:13:01 748374
## 2 LastWeekTonight Entertainment 2017-11-13 02:30:00 2418783
## 3 Rudy Mancuso Comedy 2017-11-12 14:05:24 3191434
## 4 Good Mythical Morning Entertainment 2017-11-13 06:00:04 343168
## 5 nigahiga Entertainment 2017-11-12 13:01:41 2095731
## 6 iJustine Science and Technology 2017-11-13 14:07:23 119180
## likes dislikes comment_count comments_disabled ratings_disabled
## 1 57527 2966 15954 FALSE FALSE
## 2 97185 6146 12703 FALSE FALSE
## 3 146033 5339 8181 FALSE FALSE
## 4 10172 666 2146 FALSE FALSE
## 5 132235 1989 17518 FALSE FALSE
## 6 9763 511 1434 FALSE FALSE
## video_error_or_removed publish_hour publish_when publish_wday timetotrend
## 1 FALSE 12 8am to 3pm Monday 1
## 2 FALSE 2 12am to 8am Monday 1
## 3 FALSE 14 8am to 3pm Sunday 2
## 4 FALSE 6 12am to 8am Monday 1
## 5 FALSE 13 8am to 3pm Sunday 2
## 6 FALSE 14 8am to 3pm Monday 1
## like_ratio dislike_ratio
## 1 0.07686932 0.0039632590
## 2 0.04017930 0.0025409472
## 3 0.04575780 0.0016729157
## 4 0.02964146 0.0019407404
## 5 0.06309732 0.0009490722
## 6 0.08191811 0.0042876322case1 <- yt %>%
group_by(category_id) %>%
summarise(mean_likes = mean(like_ratio)) %>%
ungroup()case1 <- case1 %>%
mutate(text_tooltip = glue("Mean Likes: {round(mean_likes, 2)}"))
case1## # A tibble: 16 × 3
## category_id mean_likes text_tooltip
## <fct> <dbl> <glue>
## 1 Autos and Vehicles 0.0176 Mean Likes: 0.02
## 2 Comedy 0.0522 Mean Likes: 0.05
## 3 Education 0.0437 Mean Likes: 0.04
## 4 Entertainment 0.0341 Mean Likes: 0.03
## 5 Film and Animation 0.0338 Mean Likes: 0.03
## 6 Gaming 0.0480 Mean Likes: 0.05
## 7 Howto and Style 0.0524 Mean Likes: 0.05
## 8 Music 0.0724 Mean Likes: 0.07
## 9 News and Politics 0.0146 Mean Likes: 0.01
## 10 Nonprofit and Activism 0.0245 Mean Likes: 0.02
## 11 People and Blogs 0.0520 Mean Likes: 0.05
## 12 Pets and Animals 0.0453 Mean Likes: 0.05
## 13 Science and Technology 0.0383 Mean Likes: 0.04
## 14 Shows 0.0322 Mean Likes: 0.03
## 15 Sports 0.0173 Mean Likes: 0.02
## 16 Travel and Events 0.0285 Mean Likes: 0.03# buat ulang ggplot
plot_case1_new <- ggplot(data = case1, aes(x = mean_likes,
y = reorder(category_id, mean_likes),
text = text_tooltip))+ # untuk menambahkan informasi tooltip yang sudah diatur
geom_col(aes(fill = mean_likes), show.legend = F)+
labs(y = "YouTube Categories", x = "Likes Ratio", title = "Average likes ratio",
subtitle = "on YouTube")+
scale_fill_gradient(low = "#fff2cd", high = "firebrick")+
theme_minimal()
ggplotly(plot_case1_new, tooltip = "text")Interpretation
- Music Category has the most likes ratio
- The least category who has the lowest like_ratio is News and Politics.
- We would like to know when is the right time to publish the video.
plot(x = yt$publish_when)
interpretation
- We can see from the visualization that from “8am to 3pm” has the most number, with 1639
- “3pm to 12am” comes after with 792
- And the least number “12am to 8am” with 552.
- After we got the information about “Public When”, now we would like to know YouTube US Trending videos, between category and Publish Hour
yt_test <- as.data.frame(table(yt$publish_when,yt$category_id))
colnames(yt_test) <- c("publish_when","category","freq")
case2 <- yt_test
plot_case2_new <- ggplot(case2, aes(x = freq,
y = reorder(category,freq),
text = glue("Frekuensi: {freq}"))) +
geom_col(aes(fill = publish_when), position = "dodge")+
labs(title = "Proportion of Youtube Trending Videos",
subtitle = "Category vs Publish Hour",
caption = "Source: Youtube Us Trending",
x = "Freq",
y = NULL,
fill = NULL)+
scale_fill_brewer(palette = "Set2")+
theme_minimal()+
theme(legend.position = "top")+
geom_text(aes(label = freq, group = publish_when),
position=position_dodge(.9))
ggplotly(plot_case2_new, tooltip = "text")Interpretation
- We can see from the vizualitation, every category shows that “8am to 3pm” has the most publish hour.
- Entertainment sits in the most trending, with 418 with the publish hour 8am to 3pm.
- The best 3 categories to make trending are “Entertainment”, “Music”. and “How to and Style”. Those three are best to post from “8 am to 3pm”
- The categories to avoid are : “Shows”, “Non-Profit and Activism”, and “Gaming”.
trending_channel <- as.data.frame(table(yt$channel_title))
head(trending_channel)## Var1
## 1 _\xa2_\xc1_\x9d
## 2 \x8b\xc4\xf8\x8b\xc4_\x8b\xc4_\x8b\xc4_ \x8b\xc4Ћĩ\x8b⦋\xc4_\x8b\xe2_ \x8c\xc9\u008c_\x8f\x8bā\x8bģ\x8b\xc4_\x8bč\x8b\xc4\xc7
## 3 \x8e\xc4_\x8e\xd9_\xa1
## 4 \x93\xf7\x81\x90\xb5_\x91⬓_\x90 Korean Englishman
## 5 12 News
## 6 1theK (\x93ݐ\x91\x8dӓ_ۓ\x9d\xab)
## Freq
## 1 1
## 2 1
## 3 1
## 4 1
## 5 1
## 6 5- which chanel is often on the Trending, at least minimum 10 times on trending
case3 <- trending_channel[trending_channel$Freq >= 10, ]
head(case3)## Var1 Freq
## 96 BBC News 13
## 173 BuzzFeedVideo 10
## 254 CNN 17
## 263 CollegeHumor 15
## 264 ColliderVideos 13
## 360 E! Entertainment 13#visualisasi
plot_case3_new <- ggplot(data = case3, mapping = aes(x = Freq,
y = reorder(Var1, Freq),
text = glue("Frekuensi: {Freq}"))) +
geom_col(mapping = aes(fill = Freq)) +
labs(title = "Top Trending Videos",
caption = "Source: Youtube Us Trending",
x = "Freq",
y = NULL,
fill = NULL)+
geom_text(mapping = aes(label = Freq),
color = "white",
size = 5,
nudge_x = -0.7) +
scale_fill_gradient(low = "orange", high = "red")
ggplotly(plot_case3_new, tooltip = "text")Interpretation
- Top 3 trending videos coming from “Refinery 29”, “The tonight show starring Jimmy Fallon”, and “Vox”.
- “Refinery 29” sits in no. 1 Trending Videos with 31 times being in trending videos
- “BuzzFeedVideo” is not as trending as today, they were in the bottom of top trending videos with only 10 times through that period.
- Which category has the most engagement between category and like ratio?
ggplot(data = yt, mapping = aes(x = like_ratio, y = category_id)) +
geom_jitter(aes(col = yt$like_ratio)) +
geom_boxplot(alpha = 0.5) +
scale_x_log10() +
geom_point() +
theme(plot.title = element_text(hjust = 0.5))## Warning: Use of `yt$like_ratio` is discouraged. Use `like_ratio` instead.## Warning: Transformation introduced infinite values in continuous x-axis
## Transformation introduced infinite values in continuous x-axis
## Transformation introduced infinite values in continuous x-axis## Warning: Removed 12 rows containing non-finite values (stat_boxplot).## Warning: Removed 12 rows containing missing values (geom_point).
Interpretation
- as we can see, Music Category has the most like ratio if we compared with another category.
- The least can be sports or nonprofit and activism.
4. Conclusion
From all graphs above, we may say some assumptions, such as : 1. Music Category holds the most promissing content based on like ratio, while News and Politics Category is not too popular amongst the other categories. 2. The best suggestion for published time is from “8am to 3pm”. 3. Music & Entertainment categories can be solution if you want to be on the top of trending list, because it has bigger chance and to be on the trending lists. 4. Although Entertainment is the number 1 based on top trending videos, but it doesnt mean it has the highest engagement rate with the viewers, the highest category based on engagement rate holds by music category.