Neural Matrix Factorization Recommendation for User Preference Prediction Based on Explicit and Implicit Feedback

Table 1

Symbol definition.

Symbol	Description
	Set of users
	Set of items
	User-item rating matrix
	Explicit feedback feature vector of user u
	Explicit feedback feature vector of item i
	User-item implicit interaction matrix
	Implicit feedback feature vector of user u
	Implicit feedback feature vector of item i
	User’s latent factor vector matrix
	Item’s latent factor vector matrix
	Activation function of the X-th layer in the neural network
	Weight matrix of the X-th layer in the neural network
	Deviation value of the X-th layer in the neural network
	Activation function of the prediction layer
	Preference prediction for the interaction between user u and item i
	Set of parameters related to the neural network
	Controlling the weight of explicit and implicit feedback in the loss calculation