Recommendation Algorithm And Application Based On Attribute Attentional Neural Matrix Factorization

The recommendation system is a measure taken by various e-commerce websites to increase the browsing volume of various types of information on the website.It provides product information and recommendations to customers,helps users find products they may be interested in,and simulates sales staff to help customers complete the purchase process.Among them,personalized recommendation is a website that builds a recommendation model based on its user profile,interest characteristics,and purchasing behavior,and recommends information and products that users are interested in to users based on this model.Matrix Factorization(MF)is one of the most intuitive and effective methods in the Recommendation System domain.It projects sparse(user,item)interactions into dense feature products which endues strong generality to the MF model.To leverage this interaction,recent works use auxiliary information of users and items.Despite effectiveness,irrationality still exists among these methods.In this work,we propose a novel model named AANMF,short for Attribute-aware Attentional Neural Matrix Factorization.AANMF combines two main parts,namely,neuralnetwork-based factorization architecture for modeling inner product and attention-mechanismbased attribute processing cell for attribute handling.AANMF can effectively solve the problem of unreasonable attribute processing.The matrix factorization structure based on the neural network restores the modeling ability of the traditional matrix factorization,and further enhances its generalization through the neural network and some improved structures.The attribute processing unit based on the attention mechanism is the focus of this work.It mainly solves the problem of the model's processing of attribute features reasonably.Extensive experiments on two real-world data sets demonstrate the robust and stronger performance of our model.Notably,we show that our model can deal with the attributes of user or item more reasonably.Further,this article also gives the entire construction and operation process detail of the recommendation system model using AANMF.