Spotify : suggesting songs based on a user’s physical parameters

Le problème est toujours l'interprétation d'un rythme cardiaque élevé. Mais ils y vont tous …

It will work by attempting to map values from these physical parameters to emotional tags. For example, a high heart rate may be mapped to ‘excitement’, while a low heart rate may be mapped to being ‘mellow’. So when songs are being recommended to a user, recommendations based on user preferences will also be filtered based on what a user’s current mood is, and what songs display similar emotional tags to what a user is currently feeling.

The idea of mood-based recommendations for music isn’t a new one. In fact, right now, Spotify does have a lot of mood-based playlists – e.g. the ‘chill playlist’. The bridge that Spotify is trying to cross by assessing a user’s current mood is having the product immediately provide the ‘emotional fix’ for a user, without the user having to do the work to understand their own mood and find the right playlist. In fact, maybe Spotify’s mood-based recommendations will begin to understand our mood and what music we need to listen to better than we do.

The danger of mood-based recommendations is the subjective judgement that needs to be made – do you suggest songs that allow a user to revel in that mood, or do you suggest songs that try to shift a user’s mood? YouTube’s recommended videos have often been criticised for leading people down topical rabbit holes that are difficult to escape from. For instance, maybe you wanted to see that video of Ben Shapiro owning a liberal snowflake, but 2 months on, you’re now being presented videos on the latest Q-Anon conspiracy. The parallel when it comes to mood may be you’re feeling depressed, Spotify recommending “sad boi / sad girl’ songs, and making it harder for you to escape that mood.

via Patent Drop : lire l’article source

Laisser un commentaire

Votre adresse e-mail ne sera pas publiée.

Ce site utilise Akismet pour réduire les indésirables. En savoir plus sur comment les données de vos commentaires sont utilisées.