TD-Take Home

Data Exploration

## Observations: 125,003
## Variables: 15
## $ tdid               <chr> "f0fd92db-69a2-4daf-a6cf-2ed33a506f28", "10...
## $ logentrytime       <dttm> 2015-05-02 19:18:00, 2015-05-02 19:29:00, ...
## $ logfileid          <int> 371391483, 371399382, 371086147, 371087588,...
## $ site               <chr> "www.youtube.com", "www.popupportal.com", "...
## $ userHourOfWeek     <dbl> 158, 158, 150, 150, 142, 152, 139, 157, 150...
## $ country            <chr> "United States", "United States", "United S...
## $ region             <chr> "Massachusetts", "Maine", "Virginia", "Virg...
## $ metro              <chr> "506", "500", "518", "518", "505", "618", "...
## $ city               <chr> "Boston", "Waterville", "Lambsburg", "Lambs...
## $ devicetype         <chr> "PC", "PC", "PC", "PC", "PC", "PC", "PC", "...
## $ osfamily           <chr> "Windows", "Windows", "Windows", "Windows",...
## $ os                 <chr> "Windows8", "Windows8", "Windows7", "Window...
## $ browser            <chr> "Chrome", "Firefox", "IE11", "IE11", "Firef...
## $ FavoriteMovieGenre <chr> "BlindedGenre1", "BlindedGenre1", "BlindedG...
## $ year               <dbl> 2015, 2015, 2015, 2015, 2015, 2015, 2015, 2...

##      tdid            logentrytime                   logfileid        
##  Length:125003      Min.   :2015-05-01 00:01:00   Min.   :369749652  
##  Class :character   1st Qu.:2015-05-08 01:52:00   1st Qu.:376315832  
##  Mode  :character   Median :2015-05-14 15:48:00   Median :382569992  
##                     Mean   :2015-05-14 16:57:32   Mean   :382351585  
##                     3rd Qu.:2015-05-21 02:30:00   3rd Qu.:388138959  
##                     Max.   :2015-05-27 22:02:00   Max.   :394043647  
##                                                                      
##      site           userHourOfWeek     country             region         
##  Length:125003      Min.   :  0.00   Length:125003      Length:125003     
##  Class :character   1st Qu.: 43.00   Class :character   Class :character  
##  Mode  :character   Median : 81.00   Mode  :character   Mode  :character  
##                     Mean   : 82.78                                        
##                     3rd Qu.:125.00                                        
##                     Max.   :167.00                                        
##                     NA's   :2627                                          
##     metro               city            devicetype       
##  Length:125003      Length:125003      Length:125003     
##  Class :character   Class :character   Class :character  
##  Mode  :character   Mode  :character   Mode  :character  
##                                                          
##                                                          
##                                                          
##                                                          
##    osfamily              os              browser         
##  Length:125003      Length:125003      Length:125003     
##  Class :character   Class :character   Class :character  
##  Mode  :character   Mode  :character   Mode  :character  
##                                                          
##                                                          
##                                                          
##                                                          
##  FavoriteMovieGenre      year     
##  Length:125003      Min.   :2015  
##  Class :character   1st Qu.:2015  
##  Mode  :character   Median :2015  
##                     Mean   :2015  
##                     3rd Qu.:2015  
##                     Max.   :2015  
## 

• userHourOfWeek has a lot of missing values

Question 1: Guess Movie Genres

## # A tibble: 5 x 2
##   FavoriteMovieGenre genre    
##   <chr>              <chr>    
## 1 BlindedGenre1      Adventure
## 2 BlindedGenre2      Drama    
## 3 BlindedGenre3      Action   
## 4 BlindedGenre4      Comedy   
## 5 BlindedGenre5      Thriller

Based on data I was able to find on the internet about most popular movie genres in the US by total revenue, I was able to determine the results shown above. In the US, Adventure made the most money by total revenue, drama the second most, etc.

## Selecting by gCt

Testing some other countries for the favorites based on data from: https://www.nyfa.edu/student-resources/12-of-the-most-popular-movie-genres-by-country/ * Brazil aligns with Action appropriately * Italy aligns with Comedy appropriately * Germany aligns with Action appropriately * There are observation that do not align with the above reference that can be attributed to sampling error

Question 2: 3 Insights

• The most popular browswer amongst the top 5 countries within the observation dataset is Chrome, followed by IE11
• PC is the most prevalent devicetype in the US
  • Top 5 states are Florida, California, Texas, Georgia, Washington
• There are 27 regions within the US that prefer genre’s other than Adventure (which prevails in the US on a whole)
  • Top 5 deviants:
  • Alabama prefers Comedy
  • Georgia prefers Drama
  • Washington, NY, NJ prefer Action

## Selecting by n()

## $Australia

## 
## $Canada

## 
## $Ireland

## 
## $`United Kingdom`

## 
## $`United States`

## Selecting by count

Question 3: Friends

• The three users who I think would be great friends are:
  • 8adb33e7-45d8-4478-ac0c-a164a306bd5a
  • 8adb33e7-45d8-4478-ac0c-a164a306bd5a
  • 8adb33e7-45d8-4478-ac0c-a164a306bd5a
• They live in the same region, use the same browser and device, and like hte same type of movie