Project Abstract

With the introduction of music streaming services, the availability of music is at an increasing scale. How can individuals be connected to music they will enjoy? The aim of a recommendation system is to produce songs to a user they will find relevant. Common techniques include content-based and collaborative filtering. Content-based filtering produces recommendations by comparing features describing a song. Collaborative filtering produces recommendations by filtering songs which similar users enjoy, which assumes similar users will enjoy similar songs. An issue facing these techniques is known as the cold start problem, where there is not enough information to recommend accurate songs to an individual. In this paper we implemented a hybrid music recommendation system reducing the implications of the cold start, while balancing novelty and diversity.

Keywords: machine learning, ai, recommendation

 Conference Details

Session: Poster Session B at Poster Stand 123

Location: Sir Stanley Clarke Auditorium at Wednesday 8th 09:00 – 12:30

Markers: Scott Yang Yang, Randell Gaya

Course: BSc Software Engineering, 3rd Year

Future Plans: I’m undecided