微型电脑应用

随着信息时代的发展，推荐系统对满足用户需求至关重要。然而，现有系统存在诸如信息稀缺和语义关系挖掘不足等问题。为此，提出了一种基于卷积神经网络的推荐系统，系统由异质信息特征融合模块、局部信息推荐模块、多标签分类的全局信息推荐模块和异质信息搜索推荐模块组成，旨在改善传统方法在搜索引擎查询推荐中的不足。实验结果显示，在MovieLens数据集上，所建系统的精确率为0.8879,召回率为0.7958,均方根误差为0.8531。在未来的研究中，可以通过进一步优化模型参数、引入更多异质信息源以及考虑用户反馈对模型进行改进。

With the development of the information age, recommendation systems are crucial in meeting users needs. However, existing systems suffer from issues such as information scarcity and insufficient mining of semantic relationships. This paper proposes a recommendation system based on convolutional neural network, which consists of heterogeneous information feature fusion module, local information recommendation module, global information recommendation module for multi label classification, and heterogeneous information search recommendation module. The system aims to improve the shortcomings of traditional methods in search engine query recommendation. The experimental results show that on the MovieLens dataset, the accuracy of the system is 0.8879, the recall is 0.7958, and the root mean square error is 0.8531. In future research, the model can be improved by further optimizing its parameters, introducing more heterogeneous information sources, and considering users feedback.