AI helps you reading Science

AI generates interpretation videos

AI extracts and analyses the key points of the paper to generate videos automatically


pub
Go Generating

AI Traceability

AI parses the academic lineage of this thesis


Master Reading Tree
Generate MRT

AI Insight

AI extracts a summary of this paper


Weibo:
ConvolutionAl Sequence Embedding Recommendation is a novel solution to top-N sequential recommendation by modeling recent actions as an “image” among time and latent dimensions and learning sequential patterns using convolutional lters

Personalized Top-N Sequential Recommendation via Convolutional Sequence Embedding.

WSDM 2018: The Eleventh ACM International Conference on Web Search and Data Mining Marina De..., (2018): 565-573

Cited by: 187|Views241
EI

Abstract

Top-N sequential recommendation models each user as a sequence of items interacted in the past and aims to predict top-N ranked items that a user will likely interact in a »near future». The order of interaction implies that sequential patterns play an important role where more recent items in a sequence have a larger impact on the next i...More

Code:

Data:

0
Introduction
  • Recommender systems have become a core technology in many applications. Most systems, e.g., top-N recommendation [9][19], recommend the items based on the user’s general preferences without paying attention to the recency of items.

    For example, some user always prefer Apple’s products to Samsung’s products.
  • To model user’s sequential patterns, the work in [17, 21] considers top-N sequential recommendation that recommends N items that a user likely interacts with in a near future
  • This problem assumes a set of users U = {u1, u2, · · · , u |U | } and a universe of items I = {i1, i2, · · · , i |I | }.
  • Unlike conventional top-N recommendation, top-N sequential recommendation models the user behavior as a sequence of items, instead of a set of items
Highlights
  • Recommender systems have become a core technology in many applications
  • To address these above limitations of existing works, we propose a ConvolutionAl Sequence Embedding Recommendation Model, or Caser for short, as a solution to top-N sequential recommendation
  • We show that Caser is a generalization of several previous models
  • As in [19, 21, 29, 32], we evaluate a model by Precision@N, Recall@N, and Mean Average Precision (MAP)
  • Caser is a novel solution to top-N sequential recommendation by modeling recent actions as an “image” among time and latent dimensions and learning sequential patterns using convolutional lters
  • Our experiments and case studies on public real life data sets suggested that Caser outperforms the state-of-the-art methods for top-N sequential recommendation
Methods
  • The proposed model, ConvolutionAl Sequence Embedding Recommendation (Caser), incorporates the Convolutional Neural Network (CNN) to learn sequential features, and Latent Factor Model (LFM) to learn user speci c features.
  • To train the CNN, for each user u, the authors extract every L successive items as input and their T items as the targets from the user’s sequence Su , shown on the left side of Figure 3.
  • This is done by sliding a window of size L + T over the user’s sequence, and each window generates a training instance for u, denoted by a triplet (u, previous L items, T items).
  • The embedding look-up operation retrieves the previous L items’ embeddings and stacks them together, resulting in a matrix E(u,t ) ∈ RL×d for user
Conclusion
  • Caser is a novel solution to top-N sequential recommendation by modeling recent actions as an “image” among time and latent dimensions and learning sequential patterns using convolutional lters.
  • This approach provides a uni ed and exible network structure for capturing many important features of sequential recommendation, i.e., point-level and union-level sequential patterns, skip behaviors, and long term user preferences.
  • The authors' experiments and case studies on public real life data sets suggested that Caser outperforms the state-of-the-art methods for top-N sequential recommendation
Tables
  • Table1: Statistics of the datasets
  • Table2: Performance comparison on the four data sets
  • Table3: MAP vs. Caser Components
Download tables as Excel
Related work
  • Conventional recommendation methods, e.g., collaborative ltering [24], matrix factorization [15, 22], and top-N recommendation [9][19], are not suitable for capturing sequential patterns because they do not model the order of actions. Early works on sequential pattern mining [1, 4] nd explicit sequential association rules based on statistical co-occurrences [17]. This approach depends on

    Embedding Look-up $% = 4 = 2 |%'(| (%,1) %

    Convolutional Layers

    Fully-connected Layers

    : h × convolution horizontal convolutional layer vertical convolutional layer

    : : × 1 max pooling ( , ) 5
Funding
  • The work of the second author is partially supported by a Discovery Grant from Natural Sciences and Engineering Research Council of Canada
Study subjects and analysis
data sets with their SI are described in Table 1: 4
We use SI to estimate the intensity of sequential signals in a data set. The four data sets with their SI are described in Table 1. MovieLens3 is the widely used movie rating data

data sets: 3
FPMC and Fossil outperform FMC on all data sets, suggesting the e ectiveness of personalization. On MovieLens, GRU4Rec achieved a performance close to Caser’s, but got a much worse performance on the other three data sets. In fact, MovieLens has more sequential signals than

data sets: 3
POP BPR FMC FPMC Fossil GRU4Rec Caser. Tmall the other three data sets, thus, the RNN-based GRU4Rec could per-. 4.2.1 Influence of Latent Dimensionality d

data sets: 3
4.2.1 Influence of Latent Dimensionality d. Figu5re 5 sho1w0s MAP20 form well on MovieLens but can easily get biased on training sets for various d while keeping the other optimal hyperparameters of the other three data sets despite the use of regularization and unchanged. On the denser MovieLens, a larger d does not always dropout as described in [8]

sparser data sets: 3
d because of over- tting. But for the other three sparser data sets, In the following studies, we examine the impact of the hyper- each model requires more latent dimensions to achieve their best parameters d, L,T one at a time by holding the remaining hyper- results. For all data sets, Caser beats the strongest baseline perforparameters at their optimal settings

data sets: 3
This is reasonable, because for a sparse data set, a higher order Markov chain tends to introduce both extra information and more noises. In most cases, Caser-2 slightly outperforms the other models on these three data sets. pastt-9 0.029

Reference
  • Rakesh Agrawal and Ramakrishnan Srikant. 1995. Mining sequential patterns. In International Conference on Data Engineering. IEEE, 3–14.
    Google ScholarLocate open access versionFindings
  • Chen Cheng, Haiqin Yang, Michael R Lyu, and Irwin King. 2013. Where You Like to Go Next: Successive Point-of-Interest Recommendation.. In International Joint Conference on Arti cial Intelligence. 2605–2611.
    Google ScholarLocate open access versionFindings
  • Eunjoon Cho, Seth A Myers, and Jure Leskovec. 2011. Friendship and mobility: user movement in location-based social networks. In International Conference on Knowledge Discovery and Data Mining. ACM, 1082–1090.
    Google ScholarLocate open access versionFindings
  • Jiawei Han, Jian Pei, and Micheline Kamber. 2011. Data mining: concepts and techniques. Elsevier.
    Google ScholarFindings
  • R. He, W.-C. Kang, and J. McAuley. 2017. Translation-based recommendation. In ACM Conference on Recommender systems.
    Google ScholarLocate open access versionFindings
  • R. He and J. McAuley. 201Fusing Similarity Models with Markov Chains for Sparse Sequential Recommendation. In International Conference on Data Mining. IEEE.
    Google ScholarLocate open access versionFindings
  • Xiangnan He, Lizi Liao, Hanwang Zhang, Liqiang Nie, Xia Hu, and Tat-Seng Chua. 201Neural collaborative ltering. In International Conference on World Wide Web. ACM, 173–182.
    Google ScholarLocate open access versionFindings
  • Balázs Hidasi, Alexandros Karatzoglou, Linas Baltrunas, and Domonkos Tikk. 2015. Session-based recommendations with recurrent neural networks. arXiv preprint arXiv:1511.06939 (2015).
    Findings
  • Yifan Hu, Yehuda Koren, and Chris Volinsky. 2008. Collaborative ltering for implicit feedback datasets. In International Conference on Data Mining. IEEE, 263–272.
    Google ScholarLocate open access versionFindings
  • Dietmar Jannach and Malte Ludewig. 2017. When Recurrent Neural Networks meet the Neighborhood for Session-Based Recommendation. In ACM Conference on Recommender systems. ACM, 306–310.
    Google ScholarLocate open access versionFindings
  • Andrej Karpathy, George Toderici, Sanketh Shetty, Thomas Leung, Rahul Sukthankar, and Li Fei-Fei. 2014. Large-scale video classi cation with convolutional neural networks. In IEEE conference on Computer Vision and Pattern Recognition. 1725–1732.
    Google ScholarLocate open access versionFindings
  • Yoon Kim. 2014. Convolutional Neural Networks for Sentence Classi cation. In Conference on Empirical Methods on Natural Language Processing. ACL, 1756– 1751.
    Google ScholarLocate open access versionFindings
  • Diederik Kingma and Jimmy Ba. 2014. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980 (2014).
    Findings
  • Yehuda Koren. 2010. Collaborative ltering with temporal dynamics. Commun. ACM 53, 4 (2010), 89–97.
    Google ScholarLocate open access versionFindings
  • Yehuda Koren, Robert Bell, and Chris Volinsky. 2009. Matrix factorization techniques for recommender systems. Computer 42, 8 (2009).
    Google ScholarLocate open access versionFindings
  • Alex Krizhevsky, Ilya Sutskever, and Geo rey E Hinton. 2012. Imagenet classi cation with deep convolutional neural networks. In Advances in Neural Information Processing Systems. 1097–1105.
    Google ScholarLocate open access versionFindings
  • Duen-Ren Liu, Chin-Hui Lai, and Wang-Jung Lee. 2009. A hybrid of sequential rules and collaborative ltering for product recommendation. Information Sciences 179, 20 (2009), 3505–3519.
    Google ScholarLocate open access versionFindings
  • Tomas Mikolov, Martin Kara át, Lukas Burget, Jan Cernocky, and Sanjeev Khudanpur. 2010. Recurrent neural network based language model.. In Interspeech, Vol. 2. 3.
    Google ScholarLocate open access versionFindings
  • Rong Pan, Yunhong Zhou, Bin Cao, Nathan N Liu, Rajan Lukose, Martin Scholz, and Qiang Yang. 2008. One-class collaborative ltering. In International Conference on Data Mining. IEEE, 502–511.
    Google ScholarLocate open access versionFindings
  • Ste en Rendle, Christoph Freudenthaler, Zeno Gantner, and Lars SchmidtThieme. 2009. BPR: Bayesian personalized ranking from implicit feedback. In Conference on Uncertainty in Arti cial Intelligence. AUAI Press, 452–461.
    Google ScholarLocate open access versionFindings
  • Ste en Rendle, Christoph Freudenthaler, and Lars Schmidt-Thieme. 2010. Factorizing personalized markov chains for next-basket recommendation. In International Conference on World Wide Web. ACM, 811–820.
    Google ScholarLocate open access versionFindings
  • Ruslan Salakhutdinov and Andriy Mnih. 2007. Probabilistic Matrix Factorization.. In Advances in Neural Information Processing Systems, Vol. 1. 2–1.
    Google ScholarLocate open access versionFindings
  • Ruslan Salakhutdinov, Andriy Mnih, and Geo rey Hinton. 2007. Restricted Boltzmann machines for collaborative ltering. In International Conference on Machine learning. ACM, 791–798.
    Google ScholarLocate open access versionFindings
  • Badrul Sarwar, George Karypis, Joseph Konstan, and John Riedl. 2001. Item-based collaborative ltering recommendation algorithms. In International Conference on World Wide Web. ACM, 285–295.
    Google ScholarLocate open access versionFindings
  • Suvash Sedhain, Aditya Krishna Menon, Scott Sanner, and Lexing Xie. 2015. Autorec: Autoencoders meet collaborative ltering. In International Conference on World Wide Web. ACM, 111–112.
    Google ScholarLocate open access versionFindings
  • Yang Song, Ali Mamdouh Elkahky, and Xiaodong He. 2016. Multi-rate deep learning for temporal recommendation. In International ACM SIGIR conference on Research and Development in Information Retrieval. ACM, 909–912.
    Google ScholarLocate open access versionFindings
  • Nitish Srivastava, Geo rey Hinton, Alex Krizhevsky, Ilya Sutskever, and Ruslan Salakhutdinov. 2014. Dropout: A Simple Way to Prevent Neural Networks from Over tting. Journal of Machine Learning Research 15, 1 (2014), 1929–1958.
    Google ScholarLocate open access versionFindings
  • Andrea Vedaldi and Karel Lenc. 2015. Matconvnet: Convolutional neural networks for matlab. In International conference on Multimedia. ACM, 689–692.
    Google ScholarLocate open access versionFindings
  • Hao Wang, Naiyan Wang, and Dit-Yan Yeung. 2015. Collaborative deep learning for recommender systems. In International Conference on Knowledge Discovery and Data Mining. ACM, 1235–1244.
    Google ScholarLocate open access versionFindings
  • Pengfei Wang, Jiafeng Guo, Yanyan Lan, Jun Xu, Shengxian Wan, and Xueqi Cheng. 2015. Learning hierarchical representation model for nextbasket recommendation. In International ACM SIGIR conference on Research and Development in Information Retrieval. ACM, 403–412.
    Google ScholarLocate open access versionFindings
  • Chao-Yuan Wu, Amr Ahmed, Alex Beutel, Alexander J. Smola, and How Jing. 2017. Recurrent Recommender Networks. In International Conference on Web Search and Data Mining. ACM, 495–503.
    Google ScholarLocate open access versionFindings
  • Yao Wu, Christopher DuBois, Alice X Zheng, and Martin Ester. 2016. Collaborative denoising auto-encoders for top-n recommender systems. In International Conference on Web Search and Data Mining. ACM, 153–162.
    Google ScholarLocate open access versionFindings
  • Quan Yuan, Gao Cong, and Aixin Sun. 2014. Graph-based point-of-interest recommendation with geographical and temporal in uences. In International Conference on Information and Knowledge Management. ACM, 659–668.
    Google ScholarLocate open access versionFindings
  • Chenyi Zhang, Ke Wang, Hongkun Yu, Jianling Sun, and Ee-Peng Lim. 2014. Latent factor transition for dynamic collaborative ltering. In SIAM International Conference on Data Mining. SIAM, 452–460.
    Google ScholarLocate open access versionFindings
  • Shenglin Zhao, Tong Zhao, Haiqin Yang, Michael R Lyu, and Irwin King. 2016. Stellar: spatial-temporal latent ranking for successive point-of-interest recommendation. In AAAI Conference on Arti cial Intelligence. AAAI Press, 315–321.
    Google ScholarLocate open access versionFindings
  • Lei Zheng, Vahid Noroozi, and Philip S. Yu. 2017. Joint Deep Modeling of Users and Items Using Reviews for Recommendation. In International Conference on Web Search and Data Mining. ACM, 425–434.
    Google ScholarLocate open access versionFindings
Author
Your rating :
0

 

Tags
Comments
数据免责声明
页面数据均来自互联网公开来源、合作出版商和通过AI技术自动分析结果,我们不对页面数据的有效性、准确性、正确性、可靠性、完整性和及时性做出任何承诺和保证。若有疑问,可以通过电子邮件方式联系我们:report@aminer.cn
小科