Assignment_6
Wubin Chen
10/18/2020
Indroduction
I scraped data from IMDb website for the top 200 most popular movies that released in last year 2019. The http link is: http://www.imdb.com/search/title?count=200&release_date=2019,2019&title_type=feature
Specifically, I scraped 8 features from this website. It includes:
- Rank: The rank of the film from 1 to 200 on the list of 200 most popular movies released in 2019.
- Title: The title of the feature film.
- Description: The description of the feature film.
- Genre: The genre of the feature film,
- Rating: The IMDb rating of the feature film.
- Votes: Votes cast in favor of the feature film.
- Director: The main director of the feature film. Note, in case of multiple directors, I’ll take only the first.
- Actor: The main actor in the feature film. Note, in case of multiple actors, I’ll take only the first.
As for the pre-processing steps after I scraped all the data, include:
- Rank: convert data to numerical format
- Title: convert title data to text
- Description: remove ‘’
- Genre: remove ‘’, extra spaces, only take the first genre if there are multiple genres
- Rating: convert it to numerical format
- Votes: remove commas and convert it to numerical
- Director: only take the first director if there are multiple directors and convert it to factors
- Actor: Only take the first actor if there are multiple actors, and convert it to factors
After pre-processing the data, save it to csv file.
Packages needed
Question 1
What type of movie has great reviews?
In order to answer this question, I select two features to analysis the impact that genre cause on review of the movie. To be more specific, I select feature rating to indicate the quality of the movie and use Genre to refer to the type of movie. And plot those two features as bar figure.
As we can observe that Action movie with good review among all the genre in terms of Rating score, because the number of Action movies are dominated in the Rating range [6,7]
In addition to that, I also analysis the number of movies for each Genre in the top 50 movies in terms of Rank. To be more specific, I select first 50 data that includes Genre and Rank, group by Genre and obtain the total number. The reason I select Rank and Genre here is because that Rank also indicates the review or quality of the movie to some degree, therefore, the number of movies in the top 50 ranked movies is also a good measure for the quality of a movie. For the table and the bar figure I created, we can obtain same conclusion as above, which Action is still the largest number of Movies in the top 50 ranked movies, it has 13 action movies, and the second is Comedy with 10.
In a short, both analysis in different aspects are concluded with same conclusion, that Action has great reviews.
| Genre | Total |
|---|---|
| Action | 13 |
| Comedy | 10 |
| Drama | 8 |
| Animation | 5 |
| Biography | 5 |
| Adventure | 4 |
| Horror | 3 |
| Crime | 1 |
| Thriller | 1 |
Question 2
What type of movie is most popular?
To solve this question, I select three features, include Genre, Rating and Votes. Then group by Genre, to average the Rating and Votes after grouped. Without loss of generality, the average votes and rating results can tell us the popularity of movies. I end up with 2 tables and two bar figures to support my result. The table 1 is sorted based on Avg_Votes in descending manner, while the table 2 is sorted based on total number of movies released in 2019 for each genre in descending manner. The figure 1 based on Genre, votes, and total number of movies, figure 2 based on Genre, Rating, and total number of movies.
In general, popularity of the movie is mainly based on how many customers watched, therefore, we conclude from the table1, which is sorted based on Avg_Votes in descending manner, that Crime, Action and Comedy are the most popular types based on the number of votes for customers. In addition to that, the average rating of those types of movies are in a good range from 6.314 to 6.811. Plus, the number of movies released in 2019 for Action and Comedy are dominated in this year among other genres. Same observation observed from figure 1.
If we look at the second table, which is sorted by the Total in descending way. (Note that total number of movies here can indicate that preference of a director). the most popular types of movies for Directors, who are willing to make those movies, are Action, Drama and Comedy. Similar findings aslo found in figure 2 to support here.
In a nutshell, we can conclude based on above two tables, that Action and Comedy are the most popular genres of movies for both customers and directors.
| Genre | Total | Avg_Rating | Avg_Votes |
|---|---|---|---|
| Crime | 9 | 6.811 | 138684 |
| Action | 49 | 6.314 | 115585 |
| Comedy | 30 | 6.567 | 97008 |
| Animation | 11 | 6.991 | 79641 |
| Adventure | 11 | 6.418 | 69163 |
| Biography | 25 | 6.896 | 52913 |
| Mystery | 1 | 6.4 | 50660 |
| Drama | 42 | 6.586 | 47836 |
| Horror | 19 | 5.837 | 43323 |
| Thriller | 1 | 4.7 | 5560 |
| Sci-Fi | 1 | 4.6 | 3395 |
| Fantasy | 1 | 6.7 | 1140 |
| Genre | Total | Avg_Rating | Avg_Votes |
|---|---|---|---|
| Action | 49 | 6.314 | 115585 |
| Drama | 42 | 6.586 | 47836 |
| Comedy | 30 | 6.567 | 97008 |
| Biography | 25 | 6.896 | 52913 |
| Horror | 19 | 5.837 | 43323 |
| Adventure | 11 | 6.418 | 69163 |
| Animation | 11 | 6.991 | 79641 |
| Crime | 9 | 6.811 | 138684 |
| Fantasy | 1 | 6.7 | 1140 |
| Mystery | 1 | 6.4 | 50660 |
| Sci-Fi | 1 | 4.6 | 3395 |
| Thriller | 1 | 4.7 | 5560 |
