基於SVD模型之變形 - WSVD 與 PSVD

以作者查詢圖書館館藏

、以作者查詢臺灣博碩士

、以作者查詢全國書目

、勘誤回報

、線上人數：76

、訪客IP：3.14.144.108

姓名

陳璞(Pu Chen) 查詢紙本館藏

畢業系所

軟體工程研究所

論文名稱

基於SVD模型之變形 - WSVD 與 PSVD
(Variants of the SVD model - WSVD and PSVD)

相關論文

★ 基於edX線上討論板社交關係之分組機制	★ 利用Kinect建置3D視覺化之Facebook互動系統
★ 利用 Kinect建置智慧型教室之評量系統	★ 基於行動裝置應用之智慧型都會區路徑規劃機制
★ 基於分析關鍵動量相關性之動態紋理轉換	★ 針對JPEG影像中隙縫修改之偵測技術
★ 基於保護影像中直線結構的細縫裁減系統	★ 建基於開放式網路社群學習環境之社群推薦機制
★ 英語作為外語的互動式情境學習環境之系統設計	★ 基於膚色保存之情感色彩轉換機制
★ A Gesture-based Presentation System for Smart Classroom using Kinect	★ 一個用於虛擬鍵盤之手勢識別框架
★ 分數冪次型灰色生成預測模型誤差分析暨電腦工具箱之研發	★ 使用慣性傳感器構建即時人體骨架動作
★ 基於多台攝影機即時三維建模	★ 基於互補度與社群網路分析於基因演算法之分組機制

檔案

[Endnote RIS 格式]

[Bibtex 格式]

[相關文章]

[文章引用]

[完整記錄]

[館藏目錄]

[檢視]

[下載]

本電子論文使用權限為同意立即開放。
已達開放權限電子全文僅授權使用者為學術研究之目的，進行個人非營利性質之檢索、閱讀、列印。
請遵守中華民國著作權法之相關規定，切勿任意重製、散佈、改作、轉貼、播送，以免觸法。

摘要(中)

隨著網際網路的發展，人們面臨越來越多的選擇，例如購物網站中商品的選擇，影音網站中該看哪些影片的選擇。推薦系統在這些網站中扮演幫人們迅速決定的重要角色。在這篇論文中，我們針對在推薦系統中的知名方法潛在因子模型 (Latent factor model)進行分析與改良，並提出了加權潛在因子模型和多項式潛在因子模型。這兩個模型分別賦予了傳統潛在因子模型權重參數和非線性的特徵組合。我們將這兩個模型對五種開放資料集進行了許多實驗，發現相較於傳統模型，這兩個模型能夠有更好地預測效果。由於我們提出的模型是基於潛在因子模型的變體，我們的模型也可以應用於其他潛在因子模型上，如SVD++模型和NMF模型。

摘要(英)

With the development of the Internet, people are faced with more and more choices, such as the choice of products in shopping websites and the choice of which videos to watch in video and audio websites. The recommendation system plays an important role in these sites to help people decide quickly. In this paper, we analyze the well known method -- the latent factor model in the recommendation system, and propose the weighted latent factor model and the polynomial latent factor model. These two models respectively give the traditional latent factor model weights and nonlinear feature combinations. We conducted many experiments on these two models for the five open data sets and found that the two models have better predictive effects than the traditional models. Since our proposed model is based on the latent factor models, our model can also be applied to other latent factor models such as SVD++ model and NMF model.

關鍵字(中)

★ 推薦系統
★ SVD模型
★ 矩陣分解模型

關鍵字(英)

★ recommender system
★ SVD model
★ matrix factorization method

論文目次

摘要 i
Abstract ii
List of Figures v
List of Tables vi
1 Introduction 1
2 Method 4
2.1 Preliminaries 4
2.1.1 Problem definition 4
2.1.2 Latent factor model 5
2.1.2.1 Basic latent factor model 5
2.1.2.2 Biased latent factor model 6
2.1.2.3 Learning algorithms 7
2.2 Weighted latent factor model 9
2.2.1 WSVD Learning algorithms 10
2.2.2 Vanishing gradient 11
2.2.3 Gradient clipping 14
2.2.4 Comparison between the WSVD model and the SVD model 14
2.3 Polynomial latent factor model 15
2.3.1 PSVD Learning algorithms 16
2.3.2 Comparison between the PSVD model and the SVD model 18
2.4 Library implementation 18
3 Experiment 21
3.1 Dataset 21
3.2 Experiment environment 22
3.3 Optimization speed of the WSVD model 22
3.4 Overall MSEs result 23
3.5 Overall MSEs result – SVD with comparable number of learnable parameters 26
3.6 Training time 26
3.7 Non-active users and long-tailed items 27
3.8 Grouped weighted latent factor model 31
4 Related work 34
4.1 Content filtering 34
iii
Contents iv
4.2 Collaborative filtering 35
4.2.1 Memory-based approach 36
4.2.2 Model-based approach 36
4.3 SVD++ 38
4.4 Factorization Machines 39
4.5 Neural Collaborative Filtering 40
4.6 Hyper-parameter optimization 40
4.6.1 Grid search 41
4.6.2 Random search 42
4.6.3 Bayesian optimization 42
5 Conclusion and future work 45
References 47

參考文獻

[1] Arora, S., Cohen, N., and Hazan, E. On the Optimization of Deep Networks:
Implicit Acceleration by Overparameterization. ArXiv e-prints (Feb. 2018).
[2] Behnel, S., Bradshaw, R., Citro, C., Dalcin, L., Seljebotn, D., and
Smith, K. Cython: The best of both worlds. Computing in Science Engineering 13,
2 (2011), 31 –39.
[3] Bergstra, J., and Bengio, Y. Random search for hyper-parameter optimization.
J. Mach. Learn. Res. 13, 1 (Feb. 2012), 281–305.
[4] Chang, C.-C., and Lin, C.-J. Libsvm: A library for support vector machines.
ACM Trans. Intell. Syst. Technol. 2, 3 (May 2011), 27:1–27:27.
[5] Cheng, H.-T., Koc, L., Harmsen, J., Shaked, T., Chandra, T., Aradhye,
H., Anderson, G., Corrado, G., Chai, W., Ispir, M., Anil, R., Haque, Z.,
Hong, L., Jain, V., Liu, X., and Shah, H. Wide & deep learning for recommender
systems. In Proceedings of the 1st Workshop on Deep Learning for Recommender
Systems (New York, NY, USA, 2016), DLRS 2016, ACM, pp. 7–10.
[6] Dooms, S., De Pessemier, T., and Martens, L. Movietweetings: a movie
rating dataset collected from twitter. In Workshop on Crowdsourcing and Human
Computation for Recommender Systems, CrowdRec at RecSys 2013 (2013).
[7] Duchi, J., Hazan, E., and Singer, Y. Adaptive subgradient methods for online
learning and stochastic optimization. J. Mach. Learn. Res. 12 (July 2011), 2121–
2159.
[8] Guo, G., Zhang, J., and Yorke-Smith, N. A novel bayesian similarity measure
for recommender systems. In Proceedings of the 23rd International Joint Conference
on Artificial Intelligence (IJCAI) (2013), pp. 2619–2625.
47
[9] Harper, F. M., and Konstan, J. A. The movielens datasets: History and context.
ACM Transactions on Interactive Intelligent Systems 5, 4 (2016), 19.
[10] He, X., Liao, L., Zhang, H., Nie, L., Hu, X., and Chua, T.-S. Neural collaborative
filtering. In Proceedings of the 26th International Conference on World Wide
Web (Republic and Canton of Geneva, Switzerland, 2017), WWW ’17, International
World Wide Web Conferences Steering Committee, pp. 173–182.
[11] Koren, Y. Factorization meets the neighborhood: A multifaceted collaborative
filtering model. In Proceedings of the 14th ACM SIGKDD International Conference
on Knowledge Discovery and Data Mining (New York, NY, USA, 2008), KDD ’08,
ACM, pp. 426–434.
[12] Linden, G., Smith, B., and York, J. Amazon.com recommendations: item-toitem
collaborative filtering. IEEE Internet Computing 7, 1 (Jan 2003), 76–80.
[13] Lops, P., de Gemmis, M., and Semeraro, G. Content-based Recommender
Systems: State of the Art and Trends. Springer US, Boston, MA, 2011, pp. 73–105.
[14] Luo, X., Zhou, M., Xia, Y., and Zhu, Q. An efficient non-negative matrixfactorization-based
approach to collaborative filtering for recommender systems.
IEEE Transactions on Industrial Informatics 10, 2 (May 2014), 1273–1284.
[15] Rendle, S. Factorization machines. In Data Mining (ICDM), 2010 IEEE 10th
International Conference on (2010), IEEE, pp. 995–1000.
[16] Snoek, J., Larochelle, H., and Adams, R. P. Practical bayesian optimization
of machine learning algorithms. In Advances in neural information processing systems
(2012), pp. 2951–2959.
[17] Takács, G., Pilászy, I., Németh, B., and Tikk, D. Matrix factorization and
neighbor based algorithms for the netflix prize problem. In Proceedings of the 2008
ACM Conference on Recommender Systems (New York, NY, USA, 2008), RecSys
’08, ACM, pp. 267–274.
48
Reference 49
[18] Wang, H., Wang, N., and Yeung, D.-Y. Collaborative deep learning for recommender
systems. In Proceedings of the 21th ACM SIGKDD International Conference
on Knowledge Discovery and Data Mining (New York, NY, USA, 2015), KDD ’15,
ACM, pp. 1235–1244.
[19] Zeiler, M. D. ADADELTA: An Adaptive Learning Rate Method. ArXiv e-prints
(Dec. 2012).
[20] Ziegler, C.-N., McNee, S. M., Konstan, J. A., and Lausen, G. Improving
recommendation lists through topic diversification. In Proceedings of the 14th International
Conference on World Wide Web (New York, NY, USA, 2005), WWW ’05,
ACM, pp. 22–32

指導教授

施國琛(Timothy K. Shih)

審核日期

2018-7-6

推文