How group recommendations work

Ordinary recommendation services work on one person. Here you enter several Last.fm usernames and pick one of two marks for each: “♥ shares their favourites” — what they already love goes into the mix — or “✦ wants new music” — they get artists they don’t play yet but probably would like.

What you get depends on how the marks combine:

• everyone ♥ — the favourites the whole group already plays, say for a road-trip playlist;
• everyone ✦ — new music meant to suit the whole group at once;
• you ✦, a friend ♥ — you discover the friend’s favourite artists you don’t know yet;
• two friends ♥ and a newcomer ✦ — the newcomer gets to know what both friends play.

For every member the service loads their Last.fm top artists over all time, the last year and the last month, picks the search starting points from them and fetches a similar-artist list for each. Every candidate is then scored separately for every person: how much they already play it relative to their own top artist, and how well it is predicted to fit them.

The finished table can be re-ranked in the browser without a second search: a slider from “liked by everyone” to “liked by anyone”, a separate weight per member, a filter on how many people an artist counts for, genre chips, and a toggle that hides what the ✦ members already play. The result link can be shared with the group.