Research and Implementation of mobile context-aware music playlist recommendation system

Abstract

Music recommendation system helps users to find the music they like quickly. This paper proposed a new way to recommend music playlist combined with the user’s context and their similar user’s music favor. Then research the technology of the mobile context-aware playlist recommendation system in detail and proposed a implement architecture of it. Introduction In recent years, with the rapid development and popularization of multimedia technology,Digital music has entered many aspects of people's lives. And in essence, it influences and changes people's multimedia consumption habits. With the continuous expansion of the scale of music database, How to effectively help users find the right music quickly and build a playlist is the main task of the music recommendation system. The traditional music recommendation system is basically divided into two categories:recommendation system based on collaborative filtering and recommendation system based on content. They simply analyze the interests of users or the melody to make a recommendation. And do not analyze the actual environment of the user. In the real world, where we are and what we do now may has a huge influence about our music flavor. Now smart phone is equipped with a variety of sensors. These sensors can be used to predict the user context. And Combined with context information can be recommended more in line with the user's current situation of the music. Mobile context-aware music playlist recommendation Context refers to the additional information that is represented in the user model, Involve physical context (for example, location, time), context (weather, light and sound intensity) and contextual information (stock quotes, sports scores), personal context (health, mood, planning activities, the social context (team activities, social activities, and who in a room), application context (email, visit the site) and system context (network connectivity status, status of the printer) and so on information. A music recommendation algorithm based on context awareness in mobile environment is proposed in this paper, is that through the intelligent terminal sensor prediction of user's context, and recommends a series of songs with the context information. Analyzing playlist title There are thousands of users to create playlists on the Internet. The playlist may is a collection about a star or may be users like collection of songs about specfic scenarios. The playlist title of semantic analysis to different playlist classification in accordance with the appropriate context. This paper uses Spotify API to obtain 1000 different users set records for the experimental data, some data format is as follows: 3rd International Conference on Materials Engineering, Manufacturing Technology and Control (ICMEMTC 2016) © 2016. The authors Published by Atlantis Press 1577 We ass similarity S behalf of th follows: 1. Split segmentati 2. Exclu 3. For e and i x us 4. m i S  Forecas Assume terminal D let D | P(Xi We including a these data, In this {walk X  , { g a d d D  to Bias the