As we’ve suggested several times in recent years, the question of whether human music recommendations are better or worse than algorithmic recommendations is likely to be a straw-man argument: the best services will use a combination of the two. Fast Company has a really interesting article on Google Play Music All Access in that regard, digging into its combination of “technical solutions like machine listening and collaborative filtering with good, old-fashioned human intuition” as it tries to understand what its users like, and what they may want to listen to next. Research scientist Doug Eck admits that there’s plenty more work to do: “It’s very hard and we haven’t solved the problem with a capital S. Nor has anybody else.” His colleague Tim Quirk, who heads the human side of Google’s music recommendations, explains why. “Algorithms can tell you what are the most popular tracks in any genre, but an algorithm might not know that “You Don’t Miss Your Water” was sort of the first classic, Southern soul ballad in that particular time signature and that it became the template for a decade’s worth of people doing the same thing,” he says. “That’s sort of arcane human knowledge.” As a primer on how Google’s approach compares to, say, Pandora or The Echo Nest, it’s a very good read.
Source: Fast Company