Design And Implementation Of Personalized Music Recommendation System Based On Context-aware

With the development of information technology,music resources have shown a massive growth trend.The personalized recommendation,as an effective solution to the problem of information overload,has received extensive attention in the music field.At present,almost all the music platforms provide personalized music recommendation services and most of their recommendation systems predict user's music preferences based on the user's listening history.However,users' short-term preferences for music are vulnerable to the influence of situational information.As a result,it is increasingly important to make reasonable use of contextual information to enhance the effectiveness of the personalized music recommendation system.At this stage,context-aware music recommendation presents the characteristics of many users,numerous music resources and rich context information,which makes the personalized music recommendation based on context-awareness face many technical challenges.Consequently,this study meets the needs of technology and market development and has certain practical significance.In view of the problem that traditional personalized music recommendation system cannot perceive the user's short-term music preference changes,this paper focuses on the combination of context-aware technology and personalized music recommendation technology.The main work is as follows.(1)In response to the problem of how to effectively obtain context information,the system in this paper uses a variety of sensors and related technology solutions to obtain user's context data actively.At the same time,according to the characteristics of big data in music recommendation,big data platform related technologies are applied to collect data from multiple sources and preprocessing is conducted to provide data support for the system.(2)Based on the collaborative filtering recommendation algorithm and classification algorithm,two context-aware music recommendation algorithms are implemented respectively to solve the problem of how to use context information for personalized recommendation.The former integrates the context similarity calculation method,so that the traditional collaborative filtering recommendation algorithm has the ability of context-awareness;the latter generates the recommendation model through the training of the classification algorithm thereby providing appropriate types of music for users in specific situations.(3)In order to generate better recommendation,a solution based on fusion classification algorithm and collaborative filtering algorithm is proposed.The recommended list generated by collaborative filtering algorithm is merged with the music type preference predicted by the classifier,which effectively improves the screening process of the candidate set,achieves the filtering of the candidate set from coarse to fine,and improves the recommendation accuracy.The experimental results indicate that this method has a certain improvement in the prediction rate.(4)In this paper,a personalized music recommendation system is designed based on context awareness.The functional test results show that the system functions are normal,the interface is beautiful and the operation is smooth.Therefore,the expected goal is achieved.The context-awareness music recommendation system presented in this paper can provide a suitable personalized music recommendation list based on the data collected from mobile phone and wearable sensor devices,on the premise of having the basic functions of the music product,which has certain practical application value.It also has certain reference significance for the research and design of other personalized music recommendation systems.