(1) Getting URL

Code 
url <- c("https://m.imdb.com/search/title/?title_type=feature&release_date=2010-01-01,2023-12-31&sort=release_date,asc&num_votes=2500,&country_of_origin=TR&count=250","https://m.imdb.com/search/title/?title_type=feature&release_date=,2009-12-31&sort=release_date,asc&num_votes=2500,&country_of_origin=TR&count=250")

(2) Preparing of DataFrame

Code 
library(tidyverse)
library(rvest)
library(stringr)
library(knitr)


convert_time <- function(time_str) {
  hours <- 0
  minutes <- 0
  
  if (grepl("h", time_str)) {
    time_components <- strsplit(time_str, "h|m")[[1]]
    if (length(time_components) >= 1) {
      hours <- as.numeric(time_components[1])
    }
    if (length(time_components) >= 2) {
      minutes <- as.numeric(time_components[2])
    }
  } else {
    minutes <- as.numeric(gsub("m", "", time_str))
  }
  
  total_minutes <- hours * 60 + minutes
  return(total_minutes)
}
#PART URL1

data_html <- read_html(url[1])
title_names1 <- data_html |> html_nodes('.ipc-title__text')
title_names1 <- html_text(title_names1)
title_names1 <- tail(head(title_names1,-1),-1)
title_names1 <- str_split(title_names1, " ", n=2)
title_names1 <- unlist(lapply(title_names1, function(x) {x[2]}))

year1 <- data_html |> html_nodes(".dli-title-metadata-item:nth-child(1)")
year1 <- as.numeric(html_text(year1))

duration1 <- data_html |> html_nodes((".dli-title-metadata-item:nth-child(2)"))
duration1 <- html_text(duration1)
duration1 <- unlist(lapply(duration1, convert_time))

rating1 <- data_html |> html_nodes(".ratingGroup--imdb-rating")
rating1 <- html_text(rating1)
rating1 <- as.numeric(str_extract(rating1, "\\d+\\.\\d+"))

votes1 <- data_html |> html_nodes(".kRnqtn")
votes1 <- html_text(votes1)
votes1 <- as.numeric(gsub("\\D", "", votes1))

#PART URL2

data_html <- read_html(url[2])
title_names2 <- data_html |> html_nodes('.ipc-title__text')
title_names2 <- html_text(title_names2)
title_names2 <- tail(head(title_names2,-1),-1)
title_names2 <- str_split(title_names2, " ", n=2)
title_names2 <- unlist(lapply(title_names2, function(x) {x[2]}))

year2 <- data_html |> html_nodes(".dli-title-metadata-item:nth-child(1)")
year2 <- as.numeric(html_text(year2))

duration2 <- data_html |> html_nodes((".dli-title-metadata-item:nth-child(2)"))
duration2 <- html_text(duration2)
duration2 <- unlist(lapply(duration2, convert_time))

rating2 <- data_html |> html_nodes(".ratingGroup--imdb-rating")
rating2 <- html_text(rating2)
rating2 <- as.numeric(str_extract(rating2, "\\d+\\.\\d+"))

votes2 <- data_html |> html_nodes(".kRnqtn")
votes2 <- html_text(votes2)
votes2 <- as.numeric(gsub("\\D", "", votes2))

Title <- c(title_names1,title_names2)
Year <- c(year1,year2)
Duration <- c(duration1,duration2)
Rating <- c(rating1,rating2)
Votes <- c(votes1,votes2)

movie_data <- data.frame(
  Title = Title,
  Year = Year,
  Duration = Duration,
  Rating = Rating,
  Votes = Votes
)

(3) Analytics of Movies

Part A

Code 
movies <- movie_data %>% 
  arrange(desc(Rating))

Top 5 movies based on ratings are shown below.

I think it wouldn’t be wrong to say that whatever exists exists in the past.

Code 
kable(
  movies %>%
    head(5),
  caption = "Top 5 Movies",
  col.names = c("Title", "Year", "Duration", "Rating", "Votes")
)
Top 5 Movies
Title Year Duration Rating Votes
Hababam Sinifi 1975 87 9.2 42520
CM101MMXI Fundamentals 2013 139 9.1 47001
Hababam Sinifi Sinifta Kaldi 1975 95 8.9 24372
Tosun Pasa 1976 90 8.9 24331
Süt Kardesler 1976 80 8.8 20891

The bottom 5 is shown below.

The new ones are obvious

Code 
kable(
  movies %>%
    tail(5),
  caption = "Bottom 5 Movies",
  col.names = c("Title", "Year", "Duration", "Rating", "Votes")
)
Bottom 5 Movies
Title Year Duration Rating Votes
466 Cumali Ceber 2 2018 100 1.2 10229
467 15/07 Safak Vakti 2021 95 1.2 20608
468 Müjde 2022 48 1.2 9920
469 Reis 2017 108 1.0 73974
470 Cumali Ceber: Allah Seni Alsin 2017 100 1.0 39268

Part B

Code 
kable(
  movie_data %>%
    filter(Title == "Ölümlü Dünya" | Title == "Yedinci Kogustaki Mucize" | Title == "Dag II"),
  caption = "My Best Three",
  col.names = c("Title", "Year", "Duration", "Rating", "Votes")
)
My Best Three
Title Year Duration Rating Votes
Dag II 2016 135 8.2 109882
Ölümlü Dünya 2018 107 7.6 30296
Yedinci Kogustaki Mucize 2019 132 8.2 54200

Part C

Code 
movie_data %>% 
  group_by(Year) %>%
  summarize(yearly_average = mean(Rating)) %>%
  ggplot(aes(x = Year, y = yearly_average)) + geom_point() +
  ggtitle("Yearly Rating Averages")

The plot clearly show that ratings decrease over the years.

Code 
ggplot(movie_data, aes(x = factor(Year))) +
  geom_bar(fill = "blue", color = "black") +
  labs(x = "Year", y = "Number of Movies") +
  ggtitle("Number of Movies Over the Years") +
  theme(legend.position = "none",
        axis.text.x = element_text(angle = 90)
  )

Code 
movie_data$Year <- as.factor(movie_data$Year)
ggplot(movie_data, aes(x = Year, y = Rating, fill = factor(Year))) +
  geom_boxplot(color = "black",width = 0.5) +
  labs(x = "Year", y = "Rating") +
  ggtitle("Box Plots of Movie Ratings Over the Years") +
  theme(legend.position = "none",
        axis.text.x = element_text(angle = 90)
  )

Part D

Code 
corr <- cor(movie_data$Votes, movie_data$Rating)
cat("Correlation between Votes and Ratings:", corr)
Correlation between Votes and Ratings: 0.1311088

The correlation between votes and ratings is weak.

Part E

Code 
corr <- cor(movie_data$Duration, movie_data$Rating)
cat("Correlation between Votes and Ratings:", corr)
Correlation between Votes and Ratings: 0.03331057

There is no correlation between duration and ratings.

(4) Turkish Movies in Top 1000

Code 
movie_data1 = movie_data
url <- c("https://m.imdb.com/search/title/?title_type=feature&sort=release_date,asc&num_votes=2500,&groups=top_1000&country_of_origin=TR&count=250")
data_html <- read_html(url)
title_names <- data_html |> html_nodes('.ipc-title__text')
title_names <- html_text(title_names)
title_names <- tail(head(title_names,-1),-1)
title_names <- str_split(title_names, " ", n=2)
title_names <- unlist(lapply(title_names, function(x) {x[2]}))

year <- data_html |> html_nodes(".dli-title-metadata-item:nth-child(1)")
year <- as.numeric(html_text(year))

top_movies <- data.frame(
  Title = title_names,
  Year = year
  )
kable (
  top_movies,
  caption = "Turkish Movies in the Top 1000",
)
Turkish Movies in the Top 1000
Title Year
Eskiya 1996
Her Sey Çok Güzel Olacak 1998
Vizontele 2001
G.O.R.A. 2004
Babam ve Oglum 2005
Nefes: Vatan Sagolsun 2009
Bir Zamanlar Anadolu’da 2011
Kis Uykusu 2014
Ayla: The Daughter of War 2017
Ahlat Agaci 2018
Yedinci Kogustaki Mucize 2019
Code 
movie_data <- movie_data %>% mutate(Year = as.numeric(Year))
top_movies <- top_movies %>% mutate(Year = as.numeric(Year))

movies_top2 <- movies %>%
  inner_join(top_movies, by = c("Title", "Year")) %>%
  arrange(desc(Rating))
kable(
  movies_top2,
  caption = "Turkish Movies in the Top 1000 Full Table",
  col.names = c("Title", "Year", "Duration", "Rating", "Votes")
)
Turkish Movies in the Top 1000 Full Table
Title Year Duration Rating Votes
Ayla: The Daughter of War 2017 125 8.3 43005
Yedinci Kogustaki Mucize 2019 132 8.2 54200
Babam ve Oglum 2005 108 8.2 91054
Eskiya 1996 128 8.1 71707
Her Sey Çok Güzel Olacak 1998 107 8.1 27129
Kis Uykusu 2014 196 8.0 54664
Ahlat Agaci 2018 188 8.0 27029
Vizontele 2001 110 8.0 38412
G.O.R.A. 2004 127 8.0 66042
Nefes: Vatan Sagolsun 2009 128 8.0 35032
Bir Zamanlar Anadolu’da 2011 157 7.8 49380
Code 
movie_data1 <- arrange(movie_data1, desc(Rating))

kable(
  head(movie_data1, 11),
  caption = "Turkish Movies in the Top 11",
  col.names = c("Title", "Year", "Duration", "Rating", "Votes")
)
Turkish Movies in the Top 11
Title Year Duration Rating Votes
Hababam Sinifi 1975 87 9.2 42520
CM101MMXI Fundamentals 2013 139 9.1 47001
Hababam Sinifi Sinifta Kaldi 1975 95 8.9 24372
Tosun Pasa 1976 90 8.9 24331
Süt Kardesler 1976 80 8.8 20891
Hababam Sinifi Uyaniyor 1976 94 8.7 20642
Saban Oglu Saban 1977 90 8.7 18537
Neseli Günler 1978 95 8.7 11811
Kibar Feyzo 1978 83 8.7 17127
Zügürt Aga 1985 101 8.7 16141
Canim Kardesim 1973 85 8.6 10099

Upon examining the table, it becomes apparent that the initial 11 highest-rated movies differ from those listed in our original data frame. A notable distinction lies in the release years of these films, with a conspicuous absence of older productions in the TOP 1000 IMDB list. It is probable that the inclusion criteria for this list prioritize movies released after a specific year.

Back to top