Personalized Recommendation System For Calligraphy Writing Machine

The interactive calligraphy writing machine with soft pen calligraphy digital creation as its core function is a new type of cultural experience equipment.The integrated calligraphy resources of this device have the characteristics of large amount of data,various types and strong homogeneity,making it difficult for experienced users efficiently searching for more interesting resources in massive resources,causing users to generate problems in using machine such as poor experience and low stickiness.Aiming at this problem,this paper proposes and implements the personalized recommendation technology of calligraphy word resources,captures user interest through mixed recommendation mode,realizes effective information filtering,and actively recommends resources that may be of interest to users.The experiment verifies the effectiveness of the proposed method on the above problems.The main research work of this paper is as follows:(1)Based on the analysis of the characteristics of calligraphy resources and the characteristics of user behavior of interactive calligraphy writing machined,the advantages and disadvantages of different personalized recommendation algorithms are compared and analyzed.(2)In order to support the content-based recommendation algorithm and the support collaborative filtering recommendation algorithm,the feature extraction technique based on VGG 16 is studied,and the vector feature is extracted from the calligraphic word resources.The user behavior based feature extraction technology on LDA theme model is studied,accomplishes feature extraction of user behavior data.(3)The hybrid recommendation strategy for the users of calligraphy characters and interactive calligraphy writing machine is studied.The content-based recommendation part maps the calligraphy characters into the same vector space through the deep learning model to construct the vector relationship between the calligraphy characters.By using the theme model method,the collaborative filtering part map the high-dimensional "user-calligraphy word" space to the low-dimensional space,and the "user-writing theme" matrix is constructed to solve the data sparseness problem;the waterfall-style hybrid recommendation model is constructed,and the filtering method is step-by-step.Get the final recommendation.(4)Developed a personalized recommendation system for calligraphy word resources for interactive calligraphy writing machine,including input and output modules,data analysis modules and recommendation engine modules,and provided personalized recommendation functions for interactive calligraphy writing machine through API interface.In this paper,facing the problem of large amount of data,large variety and homogeneity which interactive calligraphy writing machine has and derivative problems of user experience and low stickiness.We propose a personalized recommendation method for calligraphy word resources.The experiment proves the accuracy of the method.The research has important engineering guiding significance for improving the user experience of the interactive calligraphy writing machine.