Hi there, trying to leverage "similar movies" data (available through API) in my research on recommender systems. I have a couple of questions.
Has the list of similar movies for a given movie already been sorted based on relevance ?
How was this "similar set" created for each movie? To be precise, is it content-based similarity (say movies with similar attributes), or it is some sort of algorithmic recommendation based on viewership?
Thanks again for your input.
Can't find a movie or TV show? Login to create it.
Elementu hau kalifikatu edo zerrenda batera gehitzea nahi al duzu?
Ez zara kidea?
Travis Bell Erabiltzailearen Erantzuna
Otsaila 12, 2017 egunean 11:57 AM(e)tan
Hi @omidsaremi,
Similar is built by looking at and comparing plot keywords and genres. I never liked this very much which is why I swapped out that functionality on the website for the newer "recommended" system. Recommendations are powered by our user ratings and favourites. I find this typically produces better results.