General
Hi MovieLens users! We recently launched our first major revision to MovieLens since 2003! Please help us grow in the right direction by giving us thoughtful feedback on the design and functioning of the new user experience.
Ideas, questions, and bug reports are all welcome here.
89 results found
-
Export results option
I´ve like the option to export all my recomendations, my wishlist, my ratings or every list I can generate. So later I can analyze this in other software.
If this, I discovered that watched more Spielberg than I've liked... or cross later with my imdb list and verify what is diferent, what is missing.
Some data is never enough :)46 votes -
Post rotten tomatoes ratings
Post rotten tomatoes ratings
22 votesApparently RT has a workable API, so this is certainly possible for us.
-
Use IMDb plot keywords instead of tags (was: Tags)
Curently the tag system is very chaotic and if you don't change it will become worse (see listal.com -> "women i would not even look at" tags)
Why not using the official IMDB plot keywords as tags?
Users should be only able to select the pre-defined ones.
E.g. "bad" is just disturbing. Or "hot" or "boobs"...
See http://www.imdb.com/title/tt0429493/keywords?ref_=tt_stry_kw for example.
And http://www.imdb.com/Sections/Keywords/
Also all tags should be written lowercase.
10 votesThanks for the feedback. It is extremely unlikely that we will replace our tagging feature with IMDb plot keywords, but there are some other interesting ideas in here.
-
Allow arbitrary precision for predictions in stars widget
Currently, the predicted stars widgets display predicted ratings with 0.5 star precision. E.g., if the predicted rating is 3.82, it will make 3 and a half of the stars blue. It would be more useful to make three full stars and 82% of a fourth star blue.
23 votes -
User Appreciation
I think there should be some sort of rewards system for people who continue to use movielens. For example after rating say, 500 titles, could there at least be a thank you? It may seem trivial but in my opinion every one wants to feel that their input is valued and appreciated.
6 votesI think this is a really great idea.
Others: would a “thank you” email message make you feel good? Would you rather see thank you messages in the interface, or with badges/points?
-
Add a link to add to my Netflix Queue or Amazon Wish List
I would love to be able to add a movie to my Netflix queue or Amazon wish list with one click. Also would love it if ratings were synchronized. Synchronization with other systems is the future - you guys should be investigating!
9 votes -
Yet more genres would be great.
I love the increased amount of genres, but could we have even more? Maybe "Psychological" "Superhero" "Fairy Tale" and "Monster" for starters?
4 votes -
Movie Details & Options When Hovering Over Cover
It'd be great if when I hovered over a movie cover I could see a brief synopsis of the film w/additional options (i.e., Add to Wishlist, etc.), instead of having to click on the movie.
5 votes -
Remove Recent Releases From Main Page, They Can Be Found Anywhere!
It isn't difficult to find recently released movies that are popular. EVERY other recommendation site does this, you can find those movies on IMDB, Rotten Tomatoes. I don't use movielens to see whats currently playing in theaters! I mean come on you can just google that stuff.
5 votesThanks for the feedback. Other users like this feature because viewing movies in the movielens interface offers different information (e.g., predicted ratings). Let’s see if others chime in.
-
Sharable wishlist
I'd like to have some new features:
1) Sharable wishlist - just give a link to my wishlist my husband who's not registered yet so he could pick up something he'd wish to see from what I've already chosen
2) When I finally convince him to join, it would be awesome to create a recommendation list for both of us, which would take into account previous ratings of us both.
At least, it would be cool if user1 could see a wishlist or rating made for user2 but ordered by proposed ratings for user1. I believe a lot of people…1 vote -
Include more women on your team - lessen the male bias. Don't use stereotypical terms like "chick flicks" to trivialize types of movies
"Chick flicks" as a category? Inherent negative bias makes it appear that the movies are automatically shallow. I find the terminology patronizing.
2 votesI understand and fundamentally agree with your concern. However, tags are contributed by the community, not our team. We currently do not have any mechanism or policy for censoring tags. Suggestions?
FYI, you can flag tags as “bad”, which will make them less likely to show up to subsequent users.
-
increasing the Hd
HD as we know inreases the beauty and perfect picture to be observed by the clients (Viewers).This will help in increase of their need to watch them(movies).
1 voteWould you please clarify what you’re asking for? Are you asking for retina support for our images? Or, tagging movies with blu-ray?
-
List
Let us create personalized list! they are so much fun, and get people in conversations!
4 votesThanks for the idea!
-
New DVD and/or streaming releases
Need a filter to differentiate new DVD and/or stream releases from new theater releases
7 votesWe would love to do this, too. We are still considering our options, since it’s not totally clear how to get this data.
-
Quotes!
Add quotes to the movies, then it would be a great place to centralize all cinema things for everyone.
1 voteFun idea, thanks.
-
Privacy support
You could for example apply a simple data mining privacy preserving approach to your database ratings to a small extend (e.g 10%).
This might bring lower accuracy. However users mights give more ratings freely.
1 vote -
You could give the option to the users to create a trust (maybe a distrust also) network of users, such the one found in epinions
By creating a trust network perhaps the accuracy can become better, you could have more diverse recommendations and solve the cold start problem in some extend.
Happy to join and help in any way possible!
1 voteCool idea!
-
Matching Users Directly to Other Users
To complement the present system could we match users to each other directly. The idea is this, instead of predicting the ratings directly, what you would do is use a best match function of your own ratings with those of another user – you are essentially trying to find your other self. The key advantage of this method is possibly finding hidden gems – a lot of the time my predicted films are ones I know about already.
One problem with ratings is that some people’s ratings are generally cardinally higher than others which could skew this somewhat, as could…
1 voteHi Physiocrat,
The algorithm that you describe is actually the grandfather of recommendation algorithms, known as “user-user collaborative filtering”. This algorithm has lots of nice properties, especially because it is such an intuitive and interesting approach to the problem of finding new content, but its performance characteristics make it unsuitable for systems like movielens.
Your interface ideas that build on the intuition are really interesting, and there are some fun projects in there (especially combining user-user with tag information).
-
Set a movie as "favorite"
Hi, it would be great to be able to set a movie as favorite ones : sometimes I adore a movie, but I rate it lower than another one I don't like so much (because of the mood, etc). It would be nice to be able to flag our favorites movies - the one you would bring with you on a desert island.
This could eventually be achieved with lists (https://movielens.uservoice.com/forums/238501-general/suggestions/5533426-can-we-keep-a-list-and-number-of-the-movies-we-ve), but an heart icon that you click next to the stars (or something like that) would be even more nice. (like on senscritique.com)
thanks !1 voteThanks for the idea. Letterboxd also has this feature. I think lots of folks would like this.
-
prediction distribution of all unseen movies
It would be good to have the disribution on predicitions for all unseen movies. I was thinking how many movies are recommended to me with score 4.5 or 4 for example.
You guys are awesome!
Mahdi1 voteCool idea!
- Don't see your idea?