Review of Deep Learning-based Conversation Recommendation Methods
As information technology and intelligent applications continues to advance quickly,today's society is facing challenges such as data and information overload.To solve this problem,conversation-based recommendation methods have emerged.Based on deep learning,these methods use their powerful representation learning ability to more accurately predict users'short-term interests and provide personalized recommendation services.This article reviews the research progress of deep learning-based conversation recommendation methods,including methods based on convolutional neural networks,graph neural networks,attention mechanisms,multi-layer perceptrons,and hybrid models.The article summarizes the research challenges and future directions.As deep learning technology continues to make progress,conversation-based recommendation methods are expected to become increasingly important in practical applications.
NETL
NSTL
万方数据
维普
