Movie suggester how to#
Movie suggester movie#
Movie suggester series#

With an app called Flixi, you can let AI take care of your TV and movie needs.įlixi bills itself as a movie and TV personal assistant. Here are a bunch of apps and websites that will help you figure out what you should watch on Netflix. Browse Netflix’s Secret Categories (Web)įor a few bucks, Netflix puts a sea of content at your disposal but its discovery engine - oblivious to factors such as your mood, your partner’s interests - often leaves you fending for yourself to get through the endless ordeal of picking one.įortunately, there are better solutions you can rely on for Netflix recommendations. When I say that the function considers the 10 highest similarity values, I am discarding the unit value (which is easily the highest), so that the function does not return the same movie title I entered. My function, in fact, once receives the input, detects the 10 highest numbers within the row corresponding to the movie entered, gets the correspondent indexes and matches them to the movie titles series, to return the list of recommended movies.

Movie suggester series#

In order to do this, I also created a simple series of movie titles with numerical indexes, in order to match the indexes from the similarity matrix to the actual movie titles. The matrix is also symmetrical because the similarity between A and B is the same as the similarity between B and A.Īt this point, I can write the actual function that takes a movie title as input, and returns the top 10 similar movies. Let’s have a brief look at it: all the numbers on the diagonal are 1 because, of course, every movie is identical to itself. Generally, for n-dimensional vectors, the dot product can be calculated as shown below. they do not share any directions), the dot product is zero. with identical components) is equal to their squared module, while if the two are perpendicular (i.e. Therefore, the dot product between two identical vectors (i.e. The dot product between two vectors is equal to the projection of one of them on the other. Let’s think in 2D for a moment, because it’s easier to picture in our mind, and let’s refresh the concept of dot product first.

We all are familiar with vectors: they can be 2D, 3D or whatever-D. This is one of the metric that we can use when calculating similarity, between users or contents.

Movie suggester how to#

So far I have mentioned many times the word similarity, but what is it, exactly? It doesn’t really seem something we can quantify, but surprisingly it can be measured! Before jumping into a practical example of how to build a content-based system, let’s briefly review the concept of cosine similarity. There are also cluster or behavioral based for example, just to mention a couple more. These are not the only two types of existing filters, of course. Of course there are many categories we can calculate the similarity on: in the case of movies we can decide to build our own recommender system based on genre only, or maybe we want to include director, main actors and so on. If we compare this to the example above, maybe user B potentially likes dark comedy but he/she will never know, unless he/she decides to give it a try autonomously, because this filter will only keep recommending dystopian movies or similar. In this case there will be less diversity in the recommendations, but this will work either the user rates things or not. Based on what we like, the algorithm will simply pick items with similar content to recommend us. This type of filter does not involve other users if not ourselves. By looking at the items in common, this type of algorithm will basically predict the rate of a movie for a user who hasn’t watched it yet, based on the similar users’ rates. To determine whether two users are similar or not, this filter considers the movies both of them watched and how they rated them. This type of filter is based on users’ rates, and it will recommend us movies that we haven’t watched yet, but users similar to us have, and like. I will use movies as an example (because if I could, I would be watching movies/tv shows all the time), but keep in mind that this type of process can be applied for any kind of product you watch, listen to, buy, and so on. The two main types of recommender systems are either collaborative or content-based filters: these two names are pretty self-explanatory, but let’s look at a couple of examples to better understand the differences between them. We often rate products on the internet and all the preferences we express and data we share (explicitly or not), are used by recommender systems to generate, in fact, recommendations. “empty brown theater chairs” by Tyler Callahan on UnsplashĪt some point each one of us must have wondered where all the recommendations that Netflix, Amazon, Google give us, come from.