Modeling And Recommendation Of The Cultural Performance Services For The Personalized Requirements Of Audience Communities

Recent years, people pursuit for better spiritual culture, our country supportsmore strongly for cultrue industry and in this case, culture performance serviceindustry booms. The culture performance survice providers supply more abundantresourcses and the audience demand more personalized services. The selections andcombinations of adundant cultural performance service resourses to meet thedesires of audience is a problem to be detected. At the same time, more and moreInternet users especially faithful audience of culture performance. Social networksbecome hot spots for researching. There is important significance of how to minethe demands of audience in social networks and to recommend cultureperformances for them. In this paper, the modeling and recommendation of thecultural performance services for the personalized requirements of audiencecommunities aimed are put forwarded at these two points of researching mentionedabove.First of all, configurable points of services shall be identified whenconstructing customizable model to sustain the custom and recommendation ofsubsequent culture performance service. The customized models include the valuenetwork, the BPMN model, the resource model and the GRAI model. Based onthese models and the configurable node, a Bayesian Network is built with therelationship of the audience feature, the performance featrue and the ticketingservices.Secondly, we has obtained audience nodes and audience features and haspushed performance to the audience based on social network as tool andinformation transmission platform to help find potencial audience. To raise theinterest degree of audience for the performance and to enhance the effect ofperformance pushing, we propose an optimizaed slection algorighm appended withperformance feature. This algorithm selects the most likely interested performancefeature of the pushed audience and appends it in the performance information. Then, in order to discover personalized demands of different audiencecommunities, we need to divide the spread tree, which is generated when pushingperformance in soial network, into multiple communities. At the same time, pushpersonalized culture performance service for audience community. Thus, it cutsstress of providing personalized service for hundreds and thousands of audience forthe culture performance providers.To improve the satisfaction degree of culture performance service and loyaltyfor service companies, we present personalized culture performance service forpersonalized demands. Culture performance services can be divided into ticketingservices and performance services, respectively using Bayesian Network Mothodand Analytic Hierachy Process.Last but not least, we develop a culture performance service recommendationsystem to varify the feasibility of the theory above. This system contains modulesof simmulating the natural and pushing spread of performance information in socialnetwork, the discovery of audience community and its demands, the generation ofpersonalized ticketing service plan.