Netflix Movie Analysis

# Libraries
library(ggplot2)  # For data visualization
library(dplyr)     # For data manipulation
library(readr)     # For reading data files
library(stringr)   # For string manipulation
library(tidyr)     # For data tidying
library(plotly)      # For interactive plots

# Check if the file exists before loading
if(file.exists("./data/netflix_titles.csv")) {
  netflix_data <- read.csv("./data/netflix_titles.csv")
  print("File loaded successfully!")
} else {
  stop("File not found!")
}

# Display the first few rows
head(netflix_data) 

# Check for missing values in the dataset
missing_values <- colSums(is.na(netflix_data))
print(missing_values)

# Handle missing values in the 'country' column by replacing NA with "Unknown"
netflix_data <- netflix_data %>%
    mutate(country = ifelse(is.na(country), "Unknown", country))

# Convert 'date_added' to Date format and handle potential NAs
netflix_data <- netflix_data %>%
    mutate(date_added = as.Date(date_added, format = "%B %d, %Y")) %>%
    mutate(year_added = ifelse(is.na(date_added), NA, as.numeric(format(date_added, "%Y"))))

# Check if the 'year_added' column was created correctly
summary(netflix_data$year_added)

# Count the number of genres each title is listed under
netflix_data <- netflix_data %>%
    mutate(genre_count = str_count(listed_in, ",") + 1)  # Each comma indicates an additional genre

# Filter for movies and extract the duration in minutes from the 'duration' column
movie_data <- netflix_data %>%
    filter(type == "Movie") %>%
    mutate(duration = as.numeric(str_extract(duration, "\\d+")))  # Extract numeric duration

# Separate the genres into distinct rows for better analysis
movie_genres <- movie_data %>%
    separate_rows(listed_in, sep = ",\\s*") %>%
    rename(genre = listed_in)

# Convert 'duration' to numeric by removing " min" and converting to numeric
netflix_data <- netflix_data %>%
    mutate(duration = as.numeric(gsub(" min", "", duration)))  # Remove " min" and convert to numeric

# Count the number of NAs in the 'duration' column
na_count <- sum(is.na(netflix_data$duration))
cat("Number of NA values in duration:", na_count)

library(dplyr)

# Ensure 'duration' is numeric by removing any non-numeric characters (e.g., " min")
netflix_data <- netflix_data %>%
  mutate(duration = as.numeric(gsub(" min", "", duration)))

# Generate summary statistics for the dataset
summary_stats <- netflix_data %>%
  summarise(
    avg_duration = mean(duration, na.rm = TRUE),          # Average duration of all shows (in minutes)
    total_shows = n(),                                     # Total number of shows in the dataset
    avg_rating = mean(as.numeric(rating), na.rm = TRUE),    # Average rating across all shows
    total_countries = n_distinct(country),                 # Number of unique countries represented
    total_directors = n_distinct(director),                # Number of unique directors
    total_cast = sum(!is.na(cast)),                        # Total number of shows with available cast info
    total_genres = n_distinct(listed_in),                  # Number of unique genres
    .groups = "drop"                                       # Drop grouping after summarization
  )

# Display the summary statistics
summary_stats

# Filter data for titles added after 2015 and group by 'type' and 'year_added'
recent_titles <- netflix_data %>%
  filter(!is.na(year_added) & year_added > 2015) %>%
  group_by(type, year_added) %>%
  summarize(count = n(), .groups = "drop")

# Create a bar plot using ggplot2 for the trend of Movies vs. TV Shows added after 2015
plot <- ggplot(recent_titles, aes(x = year_added, y = count, fill = type)) +
  geom_col(position = "dodge") +                           # Bar plot with dodged bars for each type
  labs(
    title = "Number of Movies vs. TV Shows on Netflix (After 2015)",  # Title of the plot
    x = "Year Added",                                             # X-axis label
    y = "Count"                                                   # Y-axis label
  ) +
  theme_minimal()                                               # Minimal theme for a clean look

# Convert the ggplot2 plot to a plotly interactive plot
interactive_plot <- ggplotly(plot)

# Display the plotly interactive plot
interactive_plot

# Convert 'date_added' to 'year_added' (extracting the year from the date)
netflix_data <- netflix_data %>%
  mutate(year_added = as.numeric(format(as.Date(date_added, format = "%B %d, %Y"), "%Y")))

# Create a bar plot with ggplot2 to show the number of Netflix additions over the years
plot <- ggplot(netflix_data, aes(x = year_added)) +
  geom_bar(fill = "steelblue") +                               # Bar plot with steelblue color for the bars
  labs(
    title = "Number of Netflix Additions Over the Years",       # Title of the plot
    x = "Year Added",                                           # X-axis label: Year Added
    y = "Count"                                                 # Y-axis label: Count of titles added
  ) +
  theme_minimal()                                               # Apply minimal theme for a clean look

# Convert the ggplot2 plot to a Plotly interactive plot
interactive_plot <- ggplotly(plot)

# Display the Plotly interactive plot
interactive_plot

# Filter data and calculate average number of genres per title by year added
genre_trend <- netflix_data %>%
  filter(!is.na(year_added)) %>%
  group_by(year_added) %>%
  summarize(avg_genre_count = mean(genre_count))

# Create a line plot using ggplot2 for average genre count over time
plot <- ggplot(genre_trend, aes(x = year_added, y = avg_genre_count)) +
  geom_line(color = "blue") +                                # Line plot with blue color for the trend
  labs(
    title = "Average Number of Genres per Title on Netflix Over Time",  # Title of the plot
    x = "Year Added",                                              # X-axis label
    y = "Average Genre Count"                                       # Y-axis label
  ) +
  theme_minimal()                                                # Minimal theme for a clean and clear look

# Convert the ggplot2 plot to a plotly interactive plot
interactive_plot <- ggplotly(plot)

# Display the plotly interactive plot
interactive_plot

# Create a boxplot for movie duration by genre
plot <- ggplot(movie_genres, aes(x = genre, y = duration)) +
  geom_boxplot(fill = "tomato") +                          # Boxplot with tomato color for movie durations
  coord_flip() +                                           # Flip coordinates for better readability of genre names
  labs(
    title = "Movie Duration by Genre on Netflix",            # Title of the plot
    x = "Genre",                                            # X-axis label
    y = "Duration (Minutes)"                                 # Y-axis label
  ) +
  theme_minimal()                                           # Minimal theme for a clean and simple look

# Convert the ggplot2 plot to a plotly interactive plot
interactive_plot <- ggplotly(plot)

# Display the plotly interactive plot
interactive_plot

# Ensure you have the necessary data pre-loaded
top_countries <- netflix_data %>%
  filter(!is.na(country)) %>%   # Remove rows with missing country values
  count(country, sort = TRUE) %>% # Count the occurrences of each country
  top_n(10)  # Get the top 10 countries by content count

# Create a bar plot for the top 10 countries by content count on Netflix
plot <- ggplot(top_countries, aes(x = reorder(country, -n), y = n, fill = country)) +
  geom_bar(stat = "identity") +                              # Bar plot for content count by country
  labs(
    title = "Top 10 Countries by Content on Netflix",           # Title of the plot
    x = "Country",                                            # X-axis label
    y = "Count"                                               # Y-axis label
  ) +
  theme_minimal()                                             # Minimal theme for a clean, simple look

# Convert the ggplot2 plot to a plotly interactive plot
interactive_plot <- ggplotly(plot)

# Display the plotly interactive plot
interactive_plot