Recommended Method for Films Based on the Concept Grid of Object Threshold Variable Precision
ZHANG Guofeng
YANG Kai
Abstract:In recent years,formal concept analysis has been applied to the film recommendation system,in order to solve the problem of its recommendation accuracy is too low and the time complexity is too large,the core data structure of formal concept analysis"concept grid"is improved.First,the use of scoring information to transform the classical form background into a blurred form background.Then,for the Crisp-crisp variable-precision concept grid constructed by using this fuzzy form background,the variable-precision threshold k acting on the global is transformed into a threshold set K,and each element in the set K corresponds to a user as a separate variable-precision threshold for this user.Finally,a new Crisp-crisp variable-precision conceptual lattice structure is generated,which is named the object threshold variable-precision conceptual lattice.A series of experiments on a classi-cal film dataset illustrate the validity of this structure.In addition,the film type data in the data set is processed,combined with the nature of the pessimistic set and the optimistic set in mathematics,the film type data is converted into the film type form back-ground,the film type concept grid is generated,and finally the similarity is calculated together with the object threshold variable precision concept grid and the film type concept grid to recommend the user.Experiments have proved that this recommendation method is superior to using the object threshold variable precision concept grid alone for recommendation.
Keywords:formal conceptual analysisreferral systemobject threshold variable precision conceptual gridfilm genre con-ceptual gridmovie recommendations
Publication Date:2025-03-20
Online Publishing Date:2025-08-15(First online date of this platform, not the publication date of the document)
Pages:8( 671-677,746 )
Computer and Digital Engineering

Computer and Digital Engineering

ISTIC
ISSN:1672-9722
Year, Vol.(Issue):2025,53(3)