Music Recommendation System for Public Places Based on Sensor Network

Recently, there have been several studies about personal music recommendation system on the internet and personal computer environment, but the music recommendation system for public places has not been studied. In this paper, we study a music recommendation system based on a sensor network in public places such as highway rest places, parks, huge marts, etc. The proposed system includes five modules consisting of sensor nodes, sink nodes, host, music database, and remote control module. Sensor nodes are installed at places where people are coming and going frequently. The information sensed by sensor nodes is transmitted to the host through sink nodes using RF communication, and then the host computes the density of people from this information. In order to choose a music that best matches the current situation from our music database, several variables such as the density of people, season, weather, and time are used as main parameters. We also evaluate and analyze the accuracy of this system through an experiment environment.