Youtube EDA

library(tidyverse)
data <- read_csv("youtube.csv")

Rows: 130390 Columns: 16
── Column specification ────────────────────────────────────────────────────────────────────────────────────────
Delimiter: ","
chr  (7): video_id, title, channelId, channelTitle, tags, thumbnail_link, description
dbl  (5): categoryId, view_count, likes, dislikes, comment_count
lgl  (2): comments_disabled, ratings_disabled
dttm (2): publishedAt, trending_date

ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.

data <- unique(data)

General

Data

data

Like/Dislike ratios

head(data$likes / data$dislikes)

[1]  26.79898  52.51933 134.62215  12.52748  47.51245 103.86997

mean(head(data$likes / data$dislikes, n=1000))

[1] 67.98741

Most disliked

videos <- data %>%
  arrange(dislikes)
tail(videos, n=1)

Most Common

videos <- data %>%
  arrange(view_count)
tail(videos, n=1)

Spam checking

Comments

Videos with lots of comments are likely filled of spam.

Music videos for some reason keeps getting hit in the data, so let’s exclude it.

Big channels are showing up, lets limit the view count

This part of the report is not useful

videos <- data %>%
  filter(categoryId != 10) %>%
  filter(view_count < 100000) %>%
  filter(!comments_disabled) %>%
  select(title, comment_count) %>%
  arrange(comment_count)
tail(videos, n = 16)

Gaming

Everyone loves gaming, including me, so lets find some info about that. Gaming videos are popular, so we should find some results

Minecraft

Everyone loves MC.

videos <- data %>%
  filter(categoryId == 20) %>%
  filter(grepl("Minecraft", title))
videos