Abstract
DNR: A Unified Framework of List Ranking With Neural Networks for Recommendation
Highlights
With the rapid development of information technology, users are facing a huge number of information choices [1]– [3]
Our main contributions are as follows: 1. We present a new type of neural network architecture to fit the linear and nonlinear interaction process of users and items and devise the DNR framework for list learning based on neural networks
We have introduced two types of models: linear models represented by matrix factorization (MF) and nonlinear models represented by multi-layer perceptron (MLP)
Summary
With the rapid development of information technology, users are facing a huge number of information choices [1]– [3]. To alleviate information overload and meet the diverse needs of users, personalized recommendation systems have been developed and are beginning to play important roles in modern society [4], [5]. Jiang et al [8] presented a learning model of knowledge representation based on the self-attention mechanism, which uses the graph structure of knowledge graph and self-attention mechanism to represent entity learning features and handle complex relationship feature representation learning. These examples indicate the effectiveness of using deep learning models for collaborative recommendation
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
