融合情境感知特征的就餐推荐模型研究

Research on Restaurant Recommendation Model Combining Context-aware Feature

文章提出了一种利用深度学习技术提取用户就餐评论的情境信息的算法,利用卷积神经网络(Convolutional Neural Network,CNN)从用户对美食的评论信息中动态学习用户对美食的情景特征信息,然后将美食情境特征与初始美食潜在特征相融合构建就餐推荐模型。实验结果表明,与其他算法相比,该算法能够改善就餐推荐系统性能。

The article proposes an algorithm that utilizes deep learning technology to extract contextual information from user dining comments.The algorithm uses Convolutional Neural Network(CNN)to dynamically learn the user's situational characteristics of food from the user's food comments,and then combines the food situational characteristics with the initial food potential characteristics to build a dining recommendation model.The experimental results show that compared with other algorithms,this algorithm can improve the performance of the dining recommendation system.