Discussion 1 - Pandora Recommender

Data 612

Meaghan Burke

Prompt:

Now that we have covered basic techniques for recommender systems, choose one commercial recommender and describe how you think it works (content-based, collaborative filtering, etc). Does the technique deliver a good experience or are the recommendations off-target?

You may also choose one of the three non-personalized recommenders (below) we went over in class and describe the technique and which of the three you prefer to use.

  1. Metacritic: How We Create the Metascore Magic
  2. Rotten Tomatoes: About Rotten Tomatoes
  3. IMDB: FAQ for IMDb Ratings

Please complete the research discussion assignment in a Jupyter or R Markdown notebook. You should post the GitHub link to your research in a new discussion thread.

Pandora Overview:

Pandora is an American music streaming and automated music recommendation internet radio service powered by the Music Genome Project.The service is available via web and mobile apps. The service plays songs that have similar musical traits. [Source: Wikipedia]

How it works [Front End]:

  • Users create Radio stations based on an initial artist or song (Implicit indicator)
  • Users further refine the radio station based on a feedback loop via a Thumbs Up/Thumbs Down (Explicit indicator)
    • The feedback loop improves its recommendations

How it works [Back End]:

  • Music Genome Project: a custom created taxonomy of music developed by the founder of the company to fuel it recommendations. This taxonomy has approximately 400 attributes that describe each song in the covered universe. Unlike some recommender systems, Pandora makes selections based on songs/artists that are similarly classified in its genome to the ones selected and upvoted by the front-end user. The algorithms employed by Pandora are classified in the Content-filtering/Collaborative space by building a collection of similar content based on a group of near values.

Benefits: These associations ensure that the user is receiving consistency in content

Cost: Mis tagging by the frontend user may drastically impact the song selection and ruin the user experience