How Netflix recommends a movie or series personalised for you?
Have you ever wondered how Netflix recommends a movie/series to you? The cursory research on this topic helped me understand these things.
Todd Yellin, Netflix’s vice president of product innovation, talks about how Netflix is like a three-legged stool. The three legs are -
1. Netflix members
2. Taggers who understand the content
3. Machine learning algorithms to synthesize the data
First leg: For Netflix members, various factors are considered to estimate the likelihood of watching a particular title. A few factors are mentioned below:
- Viewing history & user title ratings
Eg: when a user gives a thumbs up to the show (Explicit data) or if a person binge-watched it(Implicit aka behavioral data)
- Information about the titles, such as their genre, categories, actors, release year
- The time of day you watch
- The devices you are watching Netflix on
- How long you watch
Second leg: The data from taggers (inhouse or freelancers), aimed at understanding the content of all shows in the form of tags
Third leg: Machine learning algorithms are used to decipher the most important aspect for users based on this data, predicting future preferences and watching habits.
PS: Yes, my Netflix feed is filled with anime