Research And Implementation Of Mini Program Recommendation Platform By Context-Aware Technology

Mini Program,as a new application form,has attracted much attention since its appearance.It is lightweight and fast,without complicated configuration.The excellent user experience promotes the rapid growth of Mini Program.However,the problem of user recommendation under massive Mini Program has become increasingly prominent.For example,users cannot quickly find the Mini Program they need when solving a temporary problem.Besides,users expect a more natural and direct way to interact with the recommendation system.Therefore,how to combine the characteristics of the contextualized service of the Mini Program and improve the recommendation interaction method to achieve the personalized recommendation of the Mini Program has become the main content of this paper.This article first conducted a literature research on the related theories and techniques of Mini Program recommendation,mainly including mobile application recommendation and context-aware recommendation.Subsequently,users,Mini Program,and context were modeled separately.The work includes constructing a feature model of the Mini Program based on the knowledge map,modeling user portraits by the fusion of multi-source tags,and completing context ontology modeling based on activity theory.Then,the hybrid recommendation strategies such as cold-start-oriented Mini Program recommendation,personalized recommendation based on the user-Mini Program-context three-dimensional association model and serialization recommendation are designed to meet users' need.Finally,a Mini Program recommendation prototype system equipped with a dialog recommendation module was designed and fulfilled.On the platform,the dialog recommendation module adopts the technology of task-based dialog,and captures the user's dialog intention to predict the user's needs and implement Mini Program recommendation.After the Mini Program recommendation prototype system is basically completed,we conducted the function tests and everything worked well.The integration of the context dimension can effectively improve the recommendation effect of Mini Program,and the interactive mode of dialog recommendation also greatly improves the user experience.