The user semantic fingerprint is at the heart of Rumo’s recommendation system. It’s a visualization of users’ history and preferences.
In Rumo’s dashboard, the user semantic fingerprint is illustrated in the section “recommendation preview”, “user recommendation” and then “profile”. It is presented in the form of bubbles. Each of them has a weight calculated based on the user’s interaction with each content and its keywords. They are divided into different categories (that you define for your catalog) : “tags”, “moods”, “character”, “format”, … You know what they say: “A picture is worth a thousand words”, and we agree! Having simple but meaningful visualizations is much better than pages and pages of user habits. Note that not only can you see it, but you also have the option to let users have access to this section of their profile by integrating the User Profile endpoint.
Find the category combinations that work best by looking at different profiles to see what interests your users. You get an overview of the top ten keywords in a user profile. You can even zoom in on each category to discover a user’s favorite actors or filmmakers.
The next step would be to let users modify these keywords and weights themselves to empower them and make them the main actors of the recommendation. By doing so, they would be able to test new keyword combinations and thus increase serendipity by themselves.
If you want this to happen, check out our public roadmap and vote for this feature to let us know it’s important to you. If it inspired you and gave you interesting ideas not mentioned yet, don’t hesitate to submit ideas!