4

I'm looking for a robust data set about movies on Bittorrent. I'd like to understand the popularity of individual titles as measured by the number of users who have a particular title available for download. I'd also like to understand the overlap between title availability. For example: 5,000 people might have Independence Day, 2,000 people might have Sleepless in Seattle, and 11 people might have both of those titles.

Is any of this data already published and freely available? If not, what would I need to do to set up some sort of a monitor to harvest the data myself? It has been widely reported that the movie studios do this to enforce their copyright so I'm assuming what I describe is possible.

|improve this question
  • 1
    Netflix does this too forbes.com/sites/timworstall/2013/09/16/… – philshem♦ May 18 '15 at 0:26
  • 3
    To avoid scraping, search Google Scholar for studies related to torrents (there're some). Authors are typically glad to share the data. – Anton Tarasenko May 22 '15 at 22:44
  • Thanks @Anton. That's a terrific suggestion not only for this question but also for future data hunting. – hughesdan Jun 22 '15 at 16:29
1

Step 1: Write a scraper to build a repository of .torrent files or magnet links. You can also use the magnet-hashes from TPB. Or convert a .torrent file to a magnet-link.

Step 2: Pass those magnet-hashes to a script that retrieves the seeder/leecher stats. An example is here.

Step 3: Repeat step 2, as many times and as often as possible.

Step 4: Occasionally update repo in step 1.

|improve this answer

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.