基于混合神经网络的协同过滤推荐模型

doi:10.3969/j.issn.0255-8297.2020.03.013

摘要/Abstract

摘要： 数据稀疏性是推荐系统中严重影响推荐结果准确性的重要因素之一.针对数据稀疏性提出了融合卷积神经网络（convolutional neural network,CNN）和降噪自编码（denoisingauto-encoder,DAE）神经网络混合的神经网络评分预测模型（convolutional-denosing autoencoder,CDAE）对用户未评分项目进行预测评分，从而解决数据稀疏性问题.首先将向量化后的用户评论数据通过卷积神经网络训练得到用户特征向量矩阵，其次将用户特征向量矩阵作为降噪自编码神经网络的初始权重，结合用户评分数据经过降噪自编码神经网络训练，得到用户-项目预测评分，然后在此基础上进行基于用户的协同过滤推荐.最后使用movielens-1M实验数据集对比验证了提出的混合神经网络协同过滤推荐（convolutional-denosing autoencoder collaborative filtering,CDAECF）模型.实验证明，所提出的CDAECF模型能够有效地结合隐性反馈和显性反馈数据，具有较高的推荐准确率.

关键词: 卷积神经网络, 降噪自编码神经网络, 协同过滤, 稀疏性

Abstract: In the recommendation system, data sparsity is one of the important factors that seriously affect the accuracy of recommendation results. Aiming at the data sparsity, this paper proposes a hybrid neural network collaborative filtering score prediction model convolutional-denosing auto-encoder (CDAE) to perform prediction scoring for solving the problem of data sparsity. The CDAE model combines a convolutional neural network (CNN) and a denoising auto-encoder (DAE) neural network. Firstly, the vectorized user review data is trained by the CNN to obtain a user feature vector matrix. Secondly, the user feature vector matrix is used as the initial weight of the DAE neural network, and the user-item prediction score is obtained by training the auto-encoder neural network in combination with user rating data. Accordingly, user-based collaborative filtering recommendations can be made. In the paper, the proposed convolutional-denosing auto-encoder collaborative filtering (CDAECF) model is experimentally verified by comparing with the experimental data set of movielens-1M. Experiment results prove that the CDAECF model can effectively combine implicit and explicit feedback data, and performs a higher recommendation accuracy rate.

Key words: convolutional neural network, denoising auto-encode neural network, collaborative filtering, sparsity

中图分类号:

TP520.60

马鑫, 吴云, 鹿泽光. 基于混合神经网络的协同过滤推荐模型[J]. 应用科学学报, 2020, 38(3): 478-487.

MA Xin, WU Yun, LU Zeguang. Collaborative Filtering Recommendation Model Based on Hybrid Neural Network[J]. Journal of Applied Sciences, 2020, 38(3): 478-487.

参考文献

[1] 王硕.基于模型的协同过滤推荐算法研究[D].北京:北京邮电大学,2019.
[2] 吴涛.推荐系统中推荐算法研究及其应用[D].北京:北京交通大学,2019.
[3] Castro J, Lu J, Zhang G Q, et al. Opinion dynamics-based group recommender systems[J]. IEEE Transactions on Systems, Man, and Cybernetics:Systems, 2018, 48(12):2394-2406.
[4] Karatzoglou A, Hidasi B. Deep learning for recommender systems[C]//In Proceedings of the Eleventh ACM Conference on Recommender Systems. ACM, 2017:396-397.
[5] Liao Y, Lu J, Liu D. News recommendation based on collaborative semantic topic models and recommendation adjustment[C]//In 2019 International Conference on Machine Learning and Cybernetics (ICMLC), 2019:1-6.
[6] Liu C, Lu T, Wang X, et al. Compositional coding for collaborative filtering[C]//In Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval. ACM, 2019:145-154.
[7] 王硕,孙光明,邹静昭,等.基于用户推荐影响度的并行协同过滤算法[J].计算机科学, 2017, 44(9):250-255, 271. Wang S, Sun G M, Zou J Z, et al. Parallel collaborative filtering recommendation algorithm based on user recommended influence[J]. Computer Science, 2017, 44(9):250-255, 271.(in Chinese)
[8] 任开旭,王玉龙,刘同存,等.融合多维语义表示的概率矩阵分解模型[J].电子学报,2019, 47(9):1848-1854. Ren K X, Wang Y L, Liu T C, et al. Probabilistic matrix factorization model incorporating multidimensional semantic representation[J]. Chinese Journal of Electronics, 2019, 47(9):1848-1854.(in Chinese)
[9] 李小浩.协同过滤推荐算法稀疏性与可扩展性问题研究[D].重庆:重庆大学,2015.
[10] 张杰,付立军,刘俊明.基于混合自编码器的协同过滤推荐算法优化[J].计算机系统应用,2019, 28(5):161-166. Zhang J, Fu L J, Liu J M. Optimization of collaborative filtering recommendation algorithm based on hybrid autoencoder[J]. Journal of Computer Applications, 2019, 28(5):161-166.(in Chinese)
[11] 黄立威,江碧涛,吕守业,等.基于深度学习的推荐系统研究综[J].计算机学报,2018, 41(7):1619-1647. Huang L W, Jiang B T, Lü S Y, et al. Survey on deep learning based recommender system[J]. Chinese Journal of Computers, 2018, 41(7):1619-1647.(in Chinese)
[12] Wang H X. An improved collaborative filtering recommendation algorithm[C]//In 2019 IEEE 4th International Conference on Big Data Analytics (ICBDA), 2019:431-435.
[13] 王美琪.基于降噪自编码器的信任感知协同过滤推荐算法研究[D].吉林:东北师范大学,2019.
[14] 李同欢.基于深度学习的推荐模型研究[D].重庆:西南大学,2019.
[15] Zhang F, Zhou W, Sun L, et al. Improvement of pearson similarity coefficient based on item frequency[C]//In 2017 International Conference on Wavelet Analysis and Pattern Recognition (ICWAPR), 2017:248-253.
[16] Xiao Y, Keung J. Improving bug localization with character-level convolutional neural network and recurrent neural network[C]//201825th Asia-Pacific Software Engineering Conference (APSEC), Nara, Japan, 2018:703-704.
[17] Zhang L, Luo T, Zhang F, et al. A recommendation model based on deep neural network[J]. IEEE Access, 2018, 6:9454-9463.
[18] Wang P, Hou H, Guo X. Collaborative filtering algorithm based on user characteristic and time weight[C]//In Proceedings of the 20198th International Conference on Software and Computer Applications. ACM, 2019:109-114.
[19] Cheng X, He D, Fang M. A hybrid collaborative filtering recommendation algorithm[C]//In Proceedings of the 2016 International Conference on Intelligent Information Processing. ACM, 2016:5:1-5:6.
[20] Gandh Si, Gandh M. Hybrid recommendation system with collaborative filtering and association rule mining using big data[C]//In 20183rd International Conference for Convergence in Technology (I2CT), 2018:1-5.
[21] 薛宇彤,陈海龙,葛苗苗.基于关联规则策略加权的Slope One算法改进[J].计算机应用研究,2019, 36(12):3605-3609. Xue Y T, Chen H L, Ge M M. Improvement of Slope One algorithm based on association rule policy weighting[J]. Application Research of Computers, 2019, 36(12):3605-3609.(in Chinese)
[22] Lu L, Yi Y, Huang F, et al. Integrating local CNN and global CNN for script identification in natural scene images[J] IEEE Access, 2019, 7:52669-52679.
[23] Huan H, Wei Z, Liang L, et al. Collaborative filtering recommendation model based on convolutional denoising auto encoder[C]//In Proceedings of the 12th Chinese Conference on Computer Supported Cooperative Work and Social Computing. ACM, 2017:64-71.
[24] Tian H, Wu L. Microblog emotional analysis based on TF-IWF weighted word2vec model[C]//In 2018 IEEE 9th International Conference on Software Engineering and Service Science (ICSESS), 2018:893-896.
[25] Shu G, Liu W, Zheng X, et al. If-CNN:image-aware inference framework for CNN with the collaboration of mobile devices and cloud[J]. IEEE Access, 2018, 6:68621-68633.