How Big Data Drives Personalized Music Recommendations

Ever wonder how music apps like Songza and Rdio figure out what music to recommend to listeners—especially listeners with unpredictably eclectic tastes that run from metal to Mozart?

The answer lies in two words: big data, according to a Forbes article. Companies like Next Big Sound scour social media platforms and ingest some 10 GB of data a day to identify trends to gain insights that help shape recommendations. “[Big data analytics] can be used to scrape data from various social media sources, parse it to get all the relevant information (artist, title, duration, genre, and so on), match it to individual preferences, and apply a recommendation algorithm,” writes Sri Raghavan, senior global product marketing manager for Teradata, a firm that helps companies “unify and analyze massive amounts of data.”

“The beauty of this lies in the simplicity of the approach. As recommendations are made to the listener, the app collects more feedback (e.g., the thumbs up or thumbs down feature in Spotify or the star feature in iTunes Radio) and recommendations can be refined to keep pace with the listener’s changing tastes.

Read the full article here.

X