Man Vs. Machine: The Curation Dilemma
Music discovery used to be simple. Our favorite radio DJs curated and made recommendations for us. They were tastemakers who introduced us to new artists. Times have changed. Some argue that the advent of collaborative filtering and data mining has made it easier to find and discover artists, while others argue that the removal of the human element has made it harder to find new music to love.
Recreating human recommendations in the digital sphere at scale is a problem we're actively solving across verticals but no one quite has the perfect formula. Where we currently stand is solving the integration of human data with machine data and algorithms to generate personalized recommendations that mirrors the nuances of human curation. This formula is the holy grail.
Panelists will explore how man and machine crawl and curate the world’s music to deliver what you want, when you want it, and why the music service that orchestrates this process best, will win.
- What mechanisms + methods are you currently using to optimize curation? Human, machine, or wizardry [Both]? Cross learning: What have we learned from each other? Humans + big data: Finding data wizards and implementing algorithms
- How do you solve predictibility with machines? Machine listening : Computers that listen and are able to pick out qualities + details Machines trying to keep up with the speed of new music
- How do you solve scale and diversity with human curation? Managing human generated data The human ability to detect various nuances The human capability of factoring musical and cultural history
- Recommending by genre is easy. How do you curate and recommend on the track level?
- The future: What are some things we can anticipate in the near future and what are your predictions?
Marc Ruxin TasetmakerX
Show me another