Design And Implementation Of Management System For The Personalized Music Recommendation

The current society is in an era of information explosion.The information is voluminous in Internet has far exceeded people can accept and handle. How to quickly and accurately location person needs information in the network becomes a problem to web site operators and customers.Music is a traditional entertainment project. The form of consumption also gradually from the tape, record and other traditional consumer to the network consumption, existence information overload problem too.If simply all music list in the web site, if customers wants to find love music they undoubtedly need to spend a lot of time to browse the massive irrelevant information, which will drowned them in the information overload problem and constantly loss. Although in order to solve this problem usually design the search engine, but this is only a negative solution, too dependent on the user input, increase website operation complexity, affect the customer experience 。 In order to solve this problem, this paper design and implementation of personalized recommendation system based on hybrid recommend music.Personalized recommendation system based on user’s music history, preference, personalized computing, active discovery registered users’ s interest, guide them to position their desired music, and even take the initiative to recommended music the user may be interested in to the user. Compared with the station search engine, the use of personalized recommendation system will allow customers to enjoy the smooth, sweet shopping experience, to establish close relationship with the registered user, let user generated dependent on site; On the other hand, it can take the initiative to promote the music web site’s exposure rate, Will change the website browsers into buyers, improve site cross selling ability, increase website traffic fundamentally.Personalized recommendation technology is a technique that essentially a kind of information filtering technology, is to use the data mining algorithm to cillect user’s personal information,and calculat the commodity that meet the user’s interest or potential interest by the appropriate algorithm. The current mainstream recommendation technology content based recommendation, collaborative filtering recommendation, the hybrid recommendation. The use of various recommendation algorithms using conditions, different scope, even if is recommended for the same music, each recommendation algorithm of recommendation results will be different. In practical application, the general will recommend several algorithms to recommend appropriate strengths fusion process of the actual formation of hybrid recommendation, can effectively improve the recommendation effect. In this paper, combined with the actual project requirements, adopt three B/S architecture, design and implement a recommender systems which according to the user’s interest and preference and browsing history, help users positioning interest music in mass music information, interference to help users avoid irrelevant information.First introduced the research background of the current research and the current research situation at home and abroad, and then give the main work in this paper of study and implementation. Secondly, learning related development technology, ensure the system programming can be carried out smoothly. Finally, introduces the overall design of process flow and database form design, then display the results of the system through the web page to and the effect of the recommendation analyzed the results.Limited personal technical ability, a web page is simple, also need to improve the business and function of the system, all these need further study in the future work.