摘要
为了提升乡村旅游体验、保护乡村文化遗产.研究引入深度神经网络和双正则化矩阵分解模型,实现了精准的个性化景点推荐,并设计了一种基于矩阵分解算法的混合推荐算法,并与虚拟现实技术相融合,最终开发出智慧乡村平台景点模拟系统,实验结果表明,研究提出的混合推荐算法准确率最高,当用户数达到800个后,该算法收敛于93.2％.此外,中年人对系统的心理可近性评价最高,好评率为94.4％.综上所述,该系统在准确性与用户满意度方面有着良好的表现.
Abstract
In order to enhance the rural tourism experience and protect rural cultural heritage.The study introduced deep neural networks and double regularized matrix factorization models to achieve accurate personalized scenic spot recommendation.A hybrid recommendation algorithm based on matrix factorization algorithm was designed and integrated with virtual reality technology to develop a smart rural platform scenic spot simulation system.The experimental results showed that the proposed hybrid recommendation algo-rithm had the highest accuracy.When the number of users reached 800,the algorithm converged to 93.2％.In addition,middle-aged people have the highest psychological accessibility evaluation of the system,with a positive rating of 94.4％.In summary,the system has performed well in terms of accuracy and user satisfaction.
基金项目
陕西省教育厅2019年度一般专项科研计划项目(19JK0682)
维普
万方数据