Transparent music recommendations
Music discovery you can understand.
We are designing a recommender that combines musical discovery with a preference profile the listener can understand and correct.
Why this question matters
A recommendation should offer something rewarding while helping the system learn about the listener. Gathering information should serve the listening experience.
What we are exploring
Conversational recommendations and feedback inform the design. The proposed system would make preferences, uncertainty, and corrections visible to the listener.
Where the work stands
The conversational prototype is not a trained recommendation system. A persistent catalog, inspectable profile, and comparative evaluation remain development goals.
Research · projects · collaboration
What are you thinking about?
Bring a question, a practical problem, or an idea you would like to explore together.