Music Search and Recommendation from Millions of Songs

Gert Lanckriet

UC San Diego

Tuesday, July 23, 2013
12: 30 p.m., Conference Room 5A

Abstract:

Advances in music production, distribution and consumption have made millions of songs available to virtually anyone on the planet, through the Internet. To allow users to retrieve the desired content from this nearly infinite pool of possibilities, algorithms for automatic music indexing and recommendation are a must.

In this talk, I will discuss two aspects of automated music analysis for music search and recommendation: i) automated music tagging for semantic retrieval (e.g., searching for ``funky jazz with male vocals''), and ii) a query-by-example paradigm for content-based music recommendation, wherein a user queries the system by providing one or more songs, and the system responds with a list of relevant or similar song recommendations (e.g., playlist generation for online radio). Finally, I will introduce our most recent research on zero-click recommendation, which leverages various ``sensor'' signals in smartphones to infer user context (activity, mood) and provide music recommendations accordingly, without requiring an active user query (zero click).

I will provide both high-level discussion and technical detail. For example, for query-by-example search, collaborative filter techniques perform well when historical data (e.g., user ratings, user playlists, etc.) is available. However, their reliance on historical data impedes performance on novel or unpopular items. To combat this problem, we rely on content-based similarity, which naturally extends to novel items, but is typically out-performed by collaborative filter methods. I will present a method for optimizing content-based similarity by learning from a sample of collaborative filter data. I will show how this algorithm may be adapted to improve recommendations if a variety of information besides musical content is available as well (e.g., music video clips, web documents and/or art work describing musical artists).

Bio:

Gert Lanckriet received a Master's degree in Electrical Engineering from the Katholieke Universiteit Leuven, Leuven, Belgium, in 2000 and the M.S. and Ph.D. degrees in Electrical Engineering and Computer Science from the University of California, Berkeley in 2001 respectively 2005. In 2005, he joined the Department of Electrical and Computer Engineering at the University of California, San Diego, where he heads the Computer Audition Lab and is a co-PI of the Distributed Health Lab. He was awarded the SIAM Optimization Prize in 2008 and is the recipient of a Hellman Fellowship, an IBM Faculty Award, an NSF CAREER Award and an Alfred P. Sloan Foundation Research Fellowship. In 2011, MIT Technology Review named him one of the 35 top young technology innovators in the world (TR35). His lab received a Yahoo! Key Scientific Challenges Award, a Qualcomm Innovation Fellowship and a Google Research Award. His research focuses on machine learning, optimization, big data analytics, and crowdsourcing, with applications in music and multimedia search and recommendation.