基于融合特征的LSTM评分预测

doi:10.3969/j.issn.1006-2475.2020.03.010

计算机与现代化

基于融合特征的LSTM评分预测

(南昌航空大学信息工程学院,江西南昌330063)

收稿日期:2019-07-19 出版日期:2020-03-24 发布日期:2020-03-30
作者简介:张尚田(1994-),男,江西上饶人,硕士研究生,研究方向:深度学习,推荐系统,E-mail： zstsmile@qq.com；通信作者：陈光,男,讲师,研究方向:信息挖掘,E-mail： cgcgcncn@163.com；邱天,女,教授,研究方向:信息挖掘。

LSTM Rating Prediction Based on Fusing Features

(School of Information Engineering, Nanchang Hangkong University, Nanchang 330063, China)

Received:2019-07-19 Online:2020-03-24 Published:2020-03-30

摘要/Abstract

摘要： 隐语义模型(LFM)能够有效地提取用户和对象的特征。本文基于LFM所提取的有效特征，提出一种基于融合特征的长短期记忆网络(LSTM)评分预测模型(F-LFM-LSTM)。首先，运用LFM模型提取用户和对象的有效特征。然后，融合用户的职业、年龄、性别标签和对象类别标签等辅助信息。最后，运用LSTM网络训练得到预测评分。通过在MovieLens100k数据集上实验表明，相比于几种得到较为广泛研究的算法，F-LFM-LSTM模型能够取得更好的评分预测效果。

关键词: 隐语义模型, F-LFM-LSTM模型, 评分预测

Abstract: The Latent Factor Model (LFM) can effectively extract the features of users and items. In this paper, based on the effective feature extracted by LFM, we propose a fusing-feature based Long Short Term Memory network (LSTM) prediction model (F-LFM-LSTM). Firstly, we employ the LFM model to extract the effective features of users and items. Then, the users occupation, age, gender label and item category label are fused. Finally, the prediction rating is obtained by training the LSTM. Experiments on MovieLens100k dataset show that, compared with several widely discussed algorithms, the F-LFM-LSTM model has higher rating prediction accuracy.

Key words: latent factor model, F-LFM-LSTM model, rating prediction

中图分类号:

TP18

张尚田，陈光，邱天. 基于融合特征的LSTM评分预测[J]. 计算机与现代化, doi: 10.3969/j.issn.1006-2475.2020.03.010.

ZHANG Shang-tian, CHEN Guang, QIU Tian. LSTM Rating Prediction Based on Fusing Features[J]. Computer and Modernization, doi: 10.3969/j.issn.1006-2475.2020.03.010.

参考文献

［1］李昂. 基于协同过滤的个性化推荐算法研究［D］. 成都:电子科技大学, 2017.
［2］成鹏,刘文斌. 基于差值矩阵分解的推荐算法［J］. 计算机与现代化, 2018(3):69-72.
［3］ WU H, YUE K,HSU C H, et al. Deviation-based neighborhood model for context-aware QoS prediction of cloud and IoT services［J］. Future Generation Computer Systems, 2017,76:550-560.
［4］刘艺,冯钧,魏童童,等. 一种改进的协同过滤推荐算法［J］. 计算机与现代化, 2017(1):1-4.
［5］范慧婷,钟春琳,龚海华. 基于隐语义模型的个性化推荐［J］. 计算机应用与软件, 2017,34(12):206-210.
［6］ YANG H X, WANG J. Bayesian hierarchical kernelized probabilistic matrix factorization［J］. Communications in Statistics-Simulation and Computation,2016,45(7):2528-2540.
［7］赵衎衎,张良富,张静,等. 因子分解机模型研究综述［J］. 软件学报, 2019,30(3):799-821.
［8］ WANG Z Q, LIANG J Y, LI R. A fusion probability matrix factorization framework for link prediction［J］. Knowledge-Based Systems, 2018,159:72-85.
［9］李晓菊,顾君忠,程洁. 基于变分循环自动编码器的协同推荐方法［J］. 计算机应用与软件, 2018,35(9):258-263.
［10］燕彩蓉,张青龙,赵雪,等. 基于广义高斯分布的贝叶斯概率矩阵分解方法［J］. 计算机研究与发展, 2016,53(12):2793-2800.
［11］LIU W X, WANG B Z, WANG D. Improved latent factor model in movie recommendation system［C］// 2018 International Conference on Intelligent Autonomous Systems. 2018:101-104.
［12］LIN R L, WANG W M. Research on the dynamic latent factor model combining attributes of user and item［C］// 2017 IEEE International Conference on Information and Automation. 2017:1082-1087.
［13］WANG E, YAO W, WANG D. Collaborative filtering recommendation algorithm optimization based on latent factor model clustering［C］// 2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery. 2017:1715-1719.
［14］王智强,梁吉业,李茹. 基于信息融合的概率矩阵分解链路预测方法［J］. 计算机研究与发展, 2019,56(2):306-318.
［15］王建芳,苗艳玲,韩鹏飞,等. 一种基于信任机制的概率矩阵分解协同过滤推荐算法［J］. 小型微型计算机系统, 2019,40(1):31-35.
［16］田保军,杨浒昀,房建东. 融合信任和基于概率矩阵分解的推荐算法［J］. 计算机应用， 2019,39(10):2834-2840.
［17］黄立威,刘艳博,李德毅. 基于深度学习的推荐系统［J/OL］. 计算机学报,(2017-11-15)［2019-07-01］. http://kns.cnki.net/kcms/detail/11.1826.TP.20171115.2242.004.html.
［18］HE X N, LIAO L Z, ZHANG H W, et al. Neural collaborative filtering［C］// Proceedings of the 26th International Conference on World Wide Web. International World Wide Web Conferences Steering Committee. 2017:173-182.
［19］CHENG H T, KOC L, HARMSEN J, et al. Wide & deep learning for recommender systems［C］// Proceedings of the 1st Workshop on Deep Learning for Recommender Systems. 2016:7-10.
［20］杨苏雁. 基于矩阵分解及深度神经网络的推荐排序学习方法研究［D］. 北京:北京交通大学, 2018.
［21］PASCANU R, MIKOLOV T, BENGIO Y. On the difficulty of training recurrent neural networks［C］// International Conference on Machine Learning. 2013:1310-1318.
［22］HOCHREITER S, SCHMIDHUBER J. LSTM can solve hard long time lag problems［C］// Proceedings of the 9th International Conference on Neural Information Processing Systems. 1997:473-479.
［23］SANCHEZ J L, SERRADILLA F, MARTINEZ E, et al. Choice of metrics used in collaborative filtering and their impact on recommender systems［C］// 2008 2nd IEEE International Conference on Digital Ecosystems and Technologies. 2008:432-436.
［24］WANG P P, QIAN Q, SHANG Z H, et al. An recommendation algorithm based on weighted slope one algorithm and user-based collaborative filtering［C］// 2016 Chinese Control and Decision Conference. 2016:2431-2434.

基于融合特征的LSTM评分预测

LSTM Rating Prediction Based on Fusing Features

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 1

编辑推荐

Metrics

本文评价