A new item-based deep network structure using a restricted Boltzmann machine for collaborative filtering

Yong-ping DU; Chang-qing YAO; Shu-hua HUO; Jing-xuan LIU

doi:10.1631/FITEE.1601732

Your Location：

Home >

Browse articles >

A new item-based deep network structure using a restricted Boltzmann machine for collaborative filtering

Regular Papers | Updated：2022-05-19

- A new item-based deep network structure using a restricted Boltzmann machine for collaborative filtering
- 一种基于受限玻尔兹曼机深度网络结构的协同过滤方法
- Frontiers of Information Technology & Electronic Engineering Vol. 18, Issue 5, Pages: 658-666(2017)
- Affiliations：
  
  1Institute of Computer Science, Beijing University of Technology, Beijing 100124, China
  2Institute of Scientific and Technical Information of China, Beijing 100038, China
- Author bio：
  
  E-mail:ypdu@bjut.edu.cn
  E-mail:yaocq@istic.ac.cn
- Funds：
- DOI：10.1631/FITEE.1601732
  CLC： TP391
- Received：22 November 2016，
  
  Accepted：07 March 2017，
  
  Published：2017-05
- Accepted：
Scan QR Code
Yong-ping DU, Chang-qing YAO, Shu-hua HUO, et al. A new item-based deep network structure using a restricted Boltzmann machine for collaborative filtering[J]. Frontiers of Information Technology & Electronic Engineering, 2017, 18(5): 658-666.
DOI：

Yong-ping DU, Chang-qing YAO, Shu-hua HUO, et al. A new item-based deep network structure using a restricted Boltzmann machine for collaborative filtering[J]. Frontiers of Information Technology & Electronic Engineering, 2017, 18(5): 658-666. DOI： 10.1631/FITEE.1601732.

摘要

协同过滤推荐算法利用历史数据进行预测推荐，在电子商务领域得到了广泛的应用，同时，数据稀疏问题依然存在。本文提出一种基于项目的受限玻尔兹曼机协同过滤算法，并采用了深度多层网络结构，有效缓解了数据稀疏问题，获取了更加有效的特征。将项目当作单独的受限玻尔兹曼机进行训练，不同的项目具有相同的权重值和偏置，在多层网络结构中，参数逐层被学习，采用带minibatch的批量梯度下降（Batch gradient descent

BGD）算法加快收敛速度，由多层玻尔兹曼机结构的网络学习到的新的特征向量在评分预测中具有更优的能力。在Movielens数据集上的实验结果表明，采用该方法的系统性能显著优于基于用户的受限玻尔兹曼机协同过滤方法，MAE与RMSE最优值分别达到了0.6424和0.7843。

Abstract

The collaborative filtering (CF) technique has been widely used recently in recommendation systems. It needs historical data to give predictions. However

the data sparsity problem still exists. We propose a new item-based restricted Boltzmann machine (RBM) approach for CF and use the deep multilayer RBM network structure

which alleviates the data sparsity problem and has excellent ability to extract features. Each item is treated as a single RBM

and different items share the same weights and biases. The parameters are learned layer by layer in the deep network. The batch gradient descent algorithm with minibatch is used to increase the convergence speed. The new feature vector discovered by the multilayer RBM network structure is very effective in predicting a rating and achieves a better result. Experimental results on the data set of MovieLens show that the item-based multilayer RBM approach achieves the best performance

with a mean absolute error of 0.6424 and a root-mean-square error of 0.7843.

关键词

Keywords

references

R Bell , , , Y Koren , , , C Volinsky . . The BellKor Solution to the Netfix Prize , , 2012 . . http://netflixprize.com/assets/GrandPrize2009_BPC_BellKor.pdf http://netflixprize.com/assets/GrandPrize2009_BPC_BellKor.pdf . .

Y Bengio . . Learning deep architectures for AI . . Found. Trends Mach. Learn. , , 2009 . . 2 ( ( 1 ): ): 1 - - 127 . . DOI: 10.1561/2200000006 http://doi.org/10.1561/2200000006 . .

J Breese , , , D Hecherman , , , C Kadie . . Empirical analysis of predictive algorithms for collaborative filtering . . 1998 . . Proc. 14th Conf. on Uncertainty in Artificial Intelligence . . 43 - - 52 . . . .

EJ Candès , , , B Recht . . Exact matrix completion via convex optimization . . Found. Comput. Math. , , 2009 . . 9 ( ( 6 ): ): 717 - - 772 . . DOI: 10.1007/s10208-009-9045-5 http://doi.org/10.1007/s10208-009-9045-5 . .

G Ference , , , M Ye , , , WC Lee . . Location recommendation for out-of-town users in location-based social networks . . 2013 . . Proc. 22nd ACM Int. Conf. on Information & Knowledge Management . . 721 - - 726 . . DOI: 10.1145/2505515.2505637 http://doi.org/10.1145/2505515.2505637 . .

A Feuerverger , , , Y He , , , S Khatri . . Statistical significance of the Netflix challenge . . Stat. Sci. , , 2012 . . 27 ( ( 2 ): ): 202 - - 231 . . DOI: 10.1214/11-STS368 http://doi.org/10.1214/11-STS368 . .

GE Hinton . . Training products of experts by minimizing contrastive divergence . . Neur. Comput. , , 2002 . . 14 ( ( 8 ): ): 1771 - - 1800 . . DOI: 10.1162/089976602760128018 http://doi.org/10.1162/089976602760128018 . .

GE Hinton , , , J Sejnowski . . Optimal perceptual inference . . 1983 . . Proc. IEEE Conf. on Computer Vision and Pattern Recognition . . 448 - - 453 . . . .

GE Hinton , , , S Osindero , , , YW Teh . . A fast learning algorithm for deep belief nets . . Neur. Comput. , , 2006 . . 18 ( ( 7 ): ): 1527 - - 1554 . . DOI: 10.1162/neco.2006.18.7.1527 http://doi.org/10.1162/neco.2006.18.7.1527 . .

Y Koren , , , R Bell , , , C Volinsky . . Matrix factorization techniques for recommender systems . . Computer , , 2009 . . 42 ( ( 8 ): ): 30 - - 37 . . DOI: 10.1109/MC.2009.263 http://doi.org/10.1109/MC.2009.263 . .

Y Li , , , M Yang , , , ZM Zhang . . Scientific articles recommendation . . 2013 . . Proc. 22nd ACM Int. Conf. on Information & Knowledge Management . . 1147 - - 1156 . . DOI: 10.1145/2505515.2505705 http://doi.org/10.1145/2505515.2505705 . .

G Linden , , , B Smith , , , J York . . Amazon . . com recommendations: item-to-item collaborative filtering. IEEE Int. Comput. , , 2003 . . 7 ( ( 1 ): ): 76 - - 80 . . DOI: 10.1109/MIC.2003.1167344 http://doi.org/10.1109/MIC.2003.1167344 . .

H Luo . . Restricted Boltzmann Machines: a Collaborative Filtering Perspective , , 2011 . . : Shanghai : Shanghai Jiao Tong University , , PhD Thesis . .

A Mohamed , , , GE Dahl , , , GE Hinton . . Acoustic modeling using deep belief networks . . IEEE Trans. Audio Speech Lang. Process. , , 2012 . . 20 ( ( 1 ): ): 14 - - 22 . . DOI: 10.1109/TASL.2011.2109382 http://doi.org/10.1109/TASL.2011.2109382 . .

T Nathanson , , , E Bitton , , , K Goldberg . . Eigentaste 5.0: constant-time adaptability in a recommender system using item clustering . . 2007 . . Proc. ACM Conf. on Recommender Systems . . 149 - - 152 . . DOI: 10.1145/1297231.1297258 http://doi.org/10.1145/1297231.1297258 . .

P Resnick , , , N Iacovou , , , M Suchak , , , 等 . . GroupLens: an open architecture for collaborative filtering of netnews . . 1994 . . Proc. ACM Conf. on Computer Supported Cooperative Work . . 175 - - 186 . . DOI: 10.1145/192844.192905 http://doi.org/10.1145/192844.192905 . .

R Salakhutdinov , , , A Mnih , , , GE Hinton . . Restricted Boltzmann machines for collaborative filtering . . 2007 . . Proc. 24th Annual Int. Conf. on Machine Learning . . 791 - - 798 . . DOI: 10.1145/1273496.1273596 http://doi.org/10.1145/1273496.1273596 . .

B Sarwar , , , G Karypis , , , J Konstan , , , 等 . . Item-based collaborative filtering recommendation algorithms . . 2001 . . Proc. ACM 10th Int. Conf. on World Wide Web . . 285 - - 295 . . DOI: 10.1145/371920.372071 http://doi.org/10.1145/371920.372071 . .

JB Schafer , , , J Konstan , , , J Riedl . . Recommender systems in E-commerce . . 1999 . . Proc. 1st ACM Conf. on Electronic Commerce . . 158 - - 166 . . DOI: 10.1145/336992.337035 http://doi.org/10.1145/336992.337035 . .

J Shi , , , J Chen , , , Z Bao . . An application study on collaborative filtering in e-commerce . . 2011 . . Int. Conf. on Service Systems and Service Management . . 1 - - 7 . . DOI: 10.1109/ICSSSM.2011.5959336 http://doi.org/10.1109/ICSSSM.2011.5959336 . .

P Smolensky . . Information processing in dynamical systems: foundations of harmony theory . . 1986 . . In: Rumellhart DE, McClelland, J.L. (Eds.), Parallel Distributed Processing: Explorations in the Microstructure of Cognition. Vol. 1: Foundations. MIT Press, Cambridge, MA, USA . . 194 - - 281 . . . .

A Töscher , , , M Jahrer . . The BigChaos Solution to the Netfix Prize , , 2008 . . http://netflixprize.com/assets/GrandPrize2009_BPC_BigChaos.pdf http://netflixprize.com/assets/GrandPrize2009_BPC_BigChaos.pdf . .

S Wang , , , J Tang , , , Y Wang . . Exploring implicit hierarchical structures for recommender systems . . 2015 . . Proc. 24th Int. Joint Conf. on Artificial Intelligence . . 1813 - - 1819 . . . .

JL Wu . . Collaborative Filtering on the Netflix Prize Dataset , , 2010 . . : Beijing : Peking University , , PhD Thesis . .

CX Zhang , , , NN Ji , , , GW Wang . . Restricted Boltzman machines . . Chin. J. Eng. Math. , , 2015 . . 32 ( ( 2 ): ): 159 - - 173 . . DOI: 10.3969/j.issn.1005-3085.2015.02.001 http://doi.org/10.3969/j.issn.1005-3085.2015.02.001 . .

M Zhang , , , J Tang , , , X Zhang , , , 等 . . Addressing cold start in recommender systems: a semi-supervised cotraining algorithm . . 2014 . . Proc. 37th Int. ACM SIGIR Conf. on Research and Development in Information Retrieval . . 73 - - 82 . . DOI: 10.1145/2600428.2609599 http://doi.org/10.1145/2600428.2609599 . .

Views

Downloads

CSCD

Alert me when the article has been cited

Submit

Tools

Publicity Resources

HAM: a deep collaborative ranking method incorporating textual information

A novel confidence estimation method for heterogeneous implicit feedback

Preference transfer model in collaborative filtering for implicit data

A microblog recommendation algorithm based on social tagging and a temporal interest evolution model*

Related Author

Teng-fei ZHOU

Chen CHEN

Tian-lei HU

Gang CHEN

Cheng-wei WANG

Peng-hua YU

Jia-qi TU

Heng ZHANG

Related Institution

The Key Laboratory of Big Data Intelligent Computing of Zhejiang Province

State Key Lab of CAD & CG, Zhejiang University

College of Computer Science and Technology, Zhejiang University

Institute of Artificial Intelligence, College of Computer Science and Technology, Zhejiang University

Health Information Center of Zhejiang Province

Chat

Address：Zhejiang University Press, 148 Tianmushan Road, Hangzhou, China Postal code：310028
Tel：+86-571-88273162 Email：fitee@zju.edu.cn
It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.

⁰