Filmaster is an application designed for film fans. Users rate, blog, and participate in forums about movies. Ratings drive the Filmaster algorithm to suggest movies that the user might like as well as ones to avoid.Show more screenshots »
Filmaster was founded in May of 2010 by Borys Musielak and is located in Warsaw Poland.
By using an algorithm, movie recommendations should be geared toward an individual’s likes and dislikes such as on Netflix. Unfortunately, during review, this was not the case. The suggestions seemed random and recommendations, for the most part, unrelated to the user’s input.
Once registered, users are invited to follow a series of steps to begin, starting with a forum introduction. The thread asks for a basic short introduction and reason why you came to Filmaster. Next, users may import ratings from IMDB or Criticker if the user has ratings on those sites.
Adding a profile is optional. Profile information may include a short bio, name, website, avatar, manage notifications – this is really important because there are TONS of notifications that are preselected. If you do not like receiving tons of email notices, definitely go to this area and uncheck those you do not wish to receive.
After creating a profile with as much or as little information as desired, the user should begin the most important step - rating movies so that the algorithm can begin to work in your favor. This is an unwieldy process as one film is offered at a time and some are foreign or obscure.
Ratings are not simple 1-10 affairs. After the general rating, users may also rate direction, screenplay, acting, special effects, editing, music, cinematography, and innovativeness. Ratings may then be reviewed if desired and added to “collection, sh*tlist, or wishlist”.
The process is slow and random. After two foreign films, during review the offerings were Forrest Gump, Home Alone, Gladiator, The Truman Show, Se7en, and The Shawshank Redemption. This process would be much faster and more rewarding with a page of films to go through rather than one page at a time. This process is slow and while it must gather data, could do it quicker with some questions and the basic rating with several on one page.
As the films progressed, they did seem to hone in on certain films based on previous responses, but my tastes are eclectic and after rating more than twenty movies, the provided films seemed more random again. Interestingly, despite multiple low ratings, animated children’s films kept popping up in the rating screen.
Short reviews appear on the user’s blog and will be added next to movie descriptions for other users.
After several reviews, Recommendations included a bizarre mix of films, many foreign, and many that did not seem to fit in with the reviewers likes at all. At this point, the algorithm did not seem terribly effective. With high scores for edgy, thriller movies, Scenes from a Marriage from 1973, Twelve Angry Men from 1957, and The Godfather from 1972 were listed along with several other older films, horror movies, and foreign films. These, while interesting, did not seem to be related to the scoring and ratings provided. Applying a filter to yield films from 2000 forward, most of the results were still odd, other than Black Swan and Inglourious Basterds.
Either the database needs more feedback, or the algorithm or screening method needs to be tweaked for this to be a really useful tool.
Users may sign in easily with Facebook, Twitter, OpenID, or create a new account. To create an account, username, password, email, and “human confirmation” (i.e. answer an arithmetic question) is required.
Filmaster is free to use.
At this point, Filmaster is not recommended. If they refine the algorithm; make the review portion more streamlined so that more general ratings take less time; and produce more relevant results, this could be a useful tool. Until then, it is not particularly effective.