Hybrid Recommender System based on Deep Learning Model

doi:10.23940/ijpe.20.01.p13.118129

Abstract

Abstract:

Mashups, which can produce enriched results using fast integrating application programming interfaces (API) and data sources, play a pivot role in building web-based and mobile applications. However, the rapidly increasing number of APIs makes it difficult to choose suitable APIs for a mashup, especially when the historical relationships between APIs and mashups are very sparse. Many researchers have sought to develop algorithms to recommend APIs to mashups. Some hybrid approaches integrate model-based collaborative filtering and auxiliary information and have achieved good prediction performance. However, there are few models that analyze the contextual information of services' descriptions. In addition, many approaches explore the context information of APIs, but few of them consider the context information of mashups. This paper presents a new model named CDHMF for recommender systems, which uses the Word2Vec technique to integrate the contextual information into a probabilistic factorization matrix (PMF). We analyze descriptions not only for APIs but also for mashups. Our experiments are performed with datasets from ProgrammableWeb. The results show that CDHMF significantly outperforms some state-of-the-art recommender systems in mashup service applications.

Key words: service recommending, deep learning, PMF, CNN, mashup, API

Chang Su and Deling Huang. Hybrid Recommender System based on Deep Learning Model [J]. Int J Performability Eng, 2020, 16(1): 118-129.

Add to citation manager EndNote|Reference Manager|ProCite|BibTeX|RefWorks

Figures/Tables 32

E = - 12 ∑ i N ∑ j M I ij (R i j - U i T V j) 2 - λ u 2 ∑ i N U i T U i - λ v 2 ∑ j M V j T V j

(1)

Figure 1.

p (R | U, V, σ 2) = ∏ i N ∏ j M N (r ij | u i T v j, σ 2) I ij

(2)

p (U | Wm, Dm, σ U 2) = ∏ i N N (u i | CNN (D m i), σ U 2) I

(3)

p (V | Wa, Da, σ V 2) = ∏ j M N (v j | CNN (D a j), σ V 2) I

(4)

p U, V, Wm, Wa R, Dm, Da, σ 2, σ U 2, σ V 2, σ Wm 2, σ Wa 2 ≈ p R U, V, σ 2 p U Wm, Dm, σ U 2 p Wm σ Wm 2 p V Wa, Da, σ V 2 p Wa σ Wa 2

(5)

Figure 2.

c = f W c × D + b

(6)

c = max c 1, c 2, ..., c l - ws + 1

(7)

d = c 1, c 2, ..., c n c

(8)

s = tanh W 2 tanh W 1 d + b 1 + b 2

(9)

max ? U, V, W p (U, V, Wm, Wa | R, Dm, Da, σ 2, σ U 2, σ V 2, σ Wm 2, σ Wa 2) = ma x U, V, W p (R | U, V, σ 2) p (U | Wm, Dm, σ U 2) p (Wm | σ Wm 2) p (V | Wa, Da, σ V 2) p (Wa | σ Wa 2)

(10)

L = ∑ i N ∑ j M I ij 2 (r i j - u i T v j) 2 + λ u 2 ∑ i N | | u i - CNN (D i) | | 2 + λ v 2 ∑ j M | | v j - CNN (D A) | | 2 + λ w ∑ k w k | | w k | | 2

(11)

u i = (V I i V T + λ U I K) - 1 V R i + λ u CNN D Mi

(12)

u i = (U I j U T + λ V I K) - 1 V R j + λ v CNN D Aj

(13)

L (W M) = λ u 2 ∑ i N | u i - CNN D A, T | 2 + λ w 2 ∑ k w k | w k | 2

(14)

? w k L (W M) = - λ U ∑ i N (u i - ? w k CNN D M i + λ W w k

(15)

L (W A) = λ v 2 ∑ j M | | v j - CNN (D A, T) | | 2 + λ w 2 ∑ k w k | | w k | | 2

(16)

? w k L (W A) = - λ V ∑ j M (v j - ? w k CNN (D A j, T j)) + λ W w k

(17)

Figure 3.

Table 1

RMSE = ∑ i, j N, M (r ij - r ? ij) 2 number ? of ? ratings

(18)

Figure 4.

Figure 5.

Recall = R A i ∩ RM A i RM A i

(19)

Recall = R (A i) ∩ RM (A i) RM (A i)

(20)

precision = 2 × recall × precision recall + precision

(21)

Figure 6.

Figure 7.

Figure 8.

Figure 9.

Figure 10.

References 23.

1.	J. Bobadilla, F. Ortega, A. Hernando,A. Gutiérrez, “Recommender Systems Survey,”Knowledge-based Systems, Vol. 46, pp. 109-132, 2013
2.	B. Cao, X. Liu, M. M. Rahman, B. Li, J. Liu,M. D. Tang, “Integrated Content and Network-based Service Clustering and Web APIs Recommendation for Mashup Development,”IEEE Transactions on Services Computing, 2017
3.	Q. Le and T. Mikolov, “Distributed Representations of Sentences and Documents,” inProceedings of International Conference on Machine Learning, pp. 1188-1196, 2014
4.	G. Ling, M. R. Lyu, King, and I. King, “Ratings Meet Reviews, a Combined Approach to Recommend,” inProceedings of the 8th ACM Conference on Recommender Systems, pp. 105-112, 2014
5.	W. Dou, X. Zhang,J. Chen, “Kasr: A Keyword-Aware Service Recommendation Method on MapReduce for Big Data Application,” IEEE Transactions on Parallel and Distributed Systems, Vol. 1, January 2014
6.	Y. Zhong, Y. Fan, W. Tan,J. Zhang, “Web Service Recommendation with Reconstructed Profile from Mashup Descriptions,” IEEE Transactions on Automation Science and Engineering, Vol. 15, No. 2, pp. 468-478, 2016
7.	L. Yao, X. Wang, Q. Z. Sheng, B. Benatallah,C. Huang, “Mashup Recommendation by Regularizing Matrix Factorization with API Co-Invocations,”IEEE Transactions on Services Computing, 2018
8.	C. Wang and D. M. Blei, “Collaborative Topic Modeling for Recommending Scientific Articles,” inProceedings of the 17th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 448-456, 2011
9.	L. Yang, “Research on Hybrid Recommendation Algorithm based on Theme Model,” Ph.D. thesis, University of Electronic Science and Technology of China, 2014
10.	A. Mnih and R. R Salakhutdinov, “Probabilistic Matrix Factorization,” in: Advances in Neural Information Processing Systems, pp. 1257-1264, 2008
11.	D. Kim, C. Park, J. Oh,H. Yu, “Deep Hybrid Recommender Systems via Exploiting Document Context and Statistics of Items,”Information Sciences, Vol. 417, pp. 72-87, 2017
12.	H. Wang, N. Wang,D. Y. Yeung, “Collaborative Deep Learning for Recommender Systems,” inProceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 1235-1244, 2015
13.	Y. Koren, R. Bell,C. Volinsky, “ Matrix Factorization Techniques for Recommender Systems,”Computer, Vol. 8, pp. 30-37, 2009
14.	A. Van den Oord, S. Dieleman, and B. Schrauwen, “Deep Content-based Music Recommendation,” in: Advances in Neural Information Processing Systems, pp. 2643-2651, 2013
15.	J. McAuley, C. Targett, Q. Shi,A. Van Den Hengel, “Image-based Recommendations on Styles and Substitutes,” inProceedings of the 38th International ACM SIGIR Conference on Research and Development in Information Retrieval, pp. 43-52, 2015
16.	J. Zhou, R. Albatal,C. Gurrin, “Applying Visual User Interest Profiles for Recommendation and Personalisation,” inProceedings of International Conference on Multimedia Modeling, pp. 361-366, 2016
17.	T. Bansal, D. Belanger,A. McCallum, “Ask the Gru: Multi-Task Learning for Deep Text Recommendations,” inProceedings of the 10th ACM Conference on Recommender Systems, pp. 107-114, 2016
18.	P. Covington, J. Adams,E. Sargin, “Deep Neural Networks for Youtube Recommendations,” inProceedings of the 10th ACM Conference on Recommender Systems, pp. 191-198, 2016
19.	G. Zanotti, M. Horvath, L. N. Barbosa, V. T. K. G. Immedisetty,J. Gemmell, “Infusing Collaborative Recommenders with Distributed Representations,” inProceedings of the 1st Workshop on Deep Learning for Recommender Systems, pp. 35-42, 2016
20.	K. Cho, B. Van Merriënboer, D. Bahdanau,Y. Bengio, “ On the Properties of Neural Machine Translation: Encoder-Decoder Approaches,” inProceedings of SSST-8, Eighth Workshop on Syntax, Semantics and Structure in Statistical Translation, pp. 103-111, 2018
21.	J. Wei, J. He, K. Chen, Y. Zhou,Z. Tang, “Collaborative Filtering and Deep Learning based Recommendation System for Cold Start Items,” Expert Systems with Applications, Vol. 29-39+69, 2017
22.	P. Vincent, H. Larochelle, I. Lajoie, Y. Bengio,P.A. Manzagol, “Stacked Denoising Autoencoders: Learning Useful Representations in a Deep Network with a Local Denoising Criterion,” Journal of Machine Learning Research, Vol. 11, pp. 3371-3408, December 2010
23.	Y. Koren, “Collaborative Filtering with Temporal Dynamics,” inProceedings of the 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 447-456, 2009

Model	Ratings	Documents	Context-info	Tags
PMF	Yes	No	No	No
CDL	Yes	Yes	No	No
R-ConMF	Yes	Yes	Yes	No
CFSMF	Yes	Yes	Yes	Yes

[1]	Lingling Wang, Wenyin Gong, and Xiang Li. Deep Convolutional Neural Network and Its Application in Image Recognition of Road Safety Projects [J]. Int J Performability Eng, 2019, 15(8): 2182-2189.
[2]	Shuang Liu, Xing Cui, Jiayi Li, Hui Yang, and Niko Lukač. Pedestrian Detection based on Faster R-CNN [J]. Int J Performability Eng, 2019, 15(7): 1792-1801.
[3]	Ying Wang, Aili Wang, Ronghui Wang, Haiyang Liu, and Yuji Iwahori. Target Tracking Algorithm based on Context-Aware Deep Feature Compression [J]. Int J Performability Eng, 2019, 15(7): 1802-1812.
[4]	Guanyu Chen, Zhihua Cai, and Xiang Li. Classification of Remote Sensing Images based on Distributed Convolutional Neural Network Model [J]. Int J Performability Eng, 2019, 15(6): 1508-1517.
[5]	Wenfang Zhao, Yong Zhou, and Wei Tang. Novel Convolution and LSTM Model for Forecasting PM2.5 Concentration [J]. Int J Performability Eng, 2019, 15(6): 1528-1537.
[6]	Sizhe Huang, Huosheng Xu, and Xuezhi Xia. Imbalanced Remote Sensing Ship Image Classification [J]. Int J Performability Eng, 2019, 15(6): 1709-1715.
[7]	Yoshinobu Tamura and Shigeru Yamada. Fault Big Data Analysis Tool based on Deep Learning [J]. Int J Performability Eng, 2019, 15(5): 1289-1296.
[8]	Yanzhou Sun, Han Wu, Yawei Zhu, Yanfang Wei, and Tieying Zhao. Fault-Section Location of Distribution Network based on Adaptive Mutation Shuffled Frog Leaping Algorithm [J]. Int J Performability Eng, 2019, 15(4): 1220-1226.
[9]	Lele Wang, Binqiang Wang, Jiangang Liu, Qiguang Miao, and Jianhui Zhang. Cuckoo-based Malware Dynamic Analysis [J]. Int J Performability Eng, 2019, 15(3): 772-781.
[10]	Haizhu Yang, Zhaoyang Jiang, Menglong Li, and Peng Zhang. Short-Term Load Forecasting based on Variational Mode Decomposition and Least Squares Support Vector Machine by Improved Artificial Fish Swarm-Shuffled Frog Jump Algorithms [J]. Int J Performability Eng, 2019, 15(12): 3117-3128.
[11]	Tao Ma, Fen Wang, Yanshan Tian, Yan Ma, and Xu Ma. A Novel Ensemble Forecasting Algorithm based on Distributed Deep Learning Network [J]. Int J Performability Eng, 2019, 15(11): 2927-2935.
[12]	Zhiyi Zhang, Chuanqi Tao, Wenhua Yang, Yuqian Zhou, and Zhiqiu Huang. A Context Model for Code and API Recommendation Systems based on Programming Onsite Data [J]. Int J Performability Eng, 2019, 15(10): 2718-2725.
[13]	Lingling Wang, Wenyin Gong, and Xiang Li. Remote Sensing Identification of Black Cotton Soil based on Deep Belief Network [J]. Int J Performability Eng, 2018, 14(11): 2820-2830.