Please wait a minute...
Big Data Mining and Analytics  2021, Vol. 4 Issue (3): 173-182    DOI: 10.26599/BDMA.2021.9020002
    
Deep Sequential Model for Anchor Recommendation on Live Streaming Platforms
Shuai Zhang1(),Hongyan Liu2,*(),Jun He1,*(),Sanpu Han3(),Xiaoyong Du1()
School of Information, Renmin University of China, Beijing 100872, China
School of Economics and Management, Tsinghua University, Beijing 100084, China
Beijing Mijing Hefeng Technology Co. Ltd., Beijing 100621, China
Download: PDF (2313 KB)      HTML  
Export: BibTeX | EndNote (RIS)      

Abstract  

Live streaming has grown rapidly in recent years, attracting increasingly more participation. As the number of online anchors is large, it is difficult for viewers to find the anchors they are interested in. Therefore, a personalized recommendation system is important for live streaming platforms. On live streaming platforms, the viewer’s and anchor’s preferences are dynamically changing over time. How to capture the user’s preference change is extensively studied in the literature, but how to model the viewer’s and anchor’s preference changes and how to learn their representations based on their preference matching are less studied. Taking these issues into consideration, in this paper, we propose a deep sequential model for live streaming recommendation. We develop a component named the multi-head related-unit in the model to capture the preference matching between anchor and viewer and extract related features for their representations. To evaluate the performance of our proposed model, we conduct experiments on real datasets, and the results show that our proposed model outperforms state-of-the-art recommendation models.



Key wordslive streaming      sequential recommendation      attention mechanism      deep learning     
Received: 27 October 2020      Published: 20 May 2021
Fund:  National Natural Science Foundation of China (NSFC)(71771131)
Corresponding Authors: Hongyan Liu,Jun He     E-mail: zhangshuai_2017@ruc.edu.cn;liuhy@sem.tsinghua.edu.cn;hejun@ruc.edu.cn;hansanpu@360.cn;duyong@ruc.edu.cn
About author: Shuai Zhang received the MS degree from Peking University, China, in 2017. He is currently pursuing the PhD degree under the guidance of Dr. Jun He at the School of Information, Renmin University of China, China. His research interests include deep learning and recommendation systems.|Hongyan Liu is a professor at the School of Economics and Management, Tsinghua University. She received the PhD degree in management science from Tsinghua University. Her current research interests include data/text mining, personalized recommendation, social computing, and medical and financial data analytics. She has published many papers in top journals, such as MISQ, ISR, INFORMS JOC, ACM TODS, ACM TOIS, and IEEE TKDE, and in top conferences, such as VLDB, ICDE, SIGKDD, ICDM, SDM, CIKM, and ICIS.|Jun He received the PhD degree in computer science from Renmin University of China, where he is now a professor and PhD supervisor. His current research interests include data mining, social network analysis, recommendation systems, and computer vision. He has published papers in many international conferences such as ACM SIGKDD, IEEE ICDM, SIAM on Data Mining, ACM CIKM, and Pacific Graphics, and in journals, such as the ACM TOIS and IEEE TKDE. He is a member of the IEEE, ACM, and AIS.|Sanpu Han is the co-founder and CTO of the Huajiao live streaming platform. He received the master degrees from Tsinghua University and Peking University in 2015 and 2010, respectively. His research interests include personalized recommendation.|Xiaoyong Du received the PhD degree from Nagoya Institute of Technology, Japan in 1997. He was an assistant professor at the Department of Intelligence and Computer Science, Nagoya Institute of Technology, from 1997 to 1999. Since 1999, he has been a professor at the School of Information, Renmin University of China. His current research interests include high performance databases, intelligent information retrieval, data mining, and the semantic web. He has published more than 100 peer-reviewed papers in journals and conferences.
Cite this article:

Shuai Zhang,Hongyan Liu,Jun He,Sanpu Han,Xiaoyong Du. Deep Sequential Model for Anchor Recommendation on Live Streaming Platforms. Big Data Mining and Analytics, 2021, 4(3): 173-182.

URL:

http://bigdata.tsinghuajournals.com/10.26599/BDMA.2021.9020002     OR     http://bigdata.tsinghuajournals.com/Y2021/V4/I3/173

Fig. 1 Architecture of the DARM.
Fig. 2 Architecture of the ADARM.
Fig. 3 Architecture of the multi-head related-unit.
ViewerAnchorInteractionAverage watching sequence lengthAverage broadcasting sequence lengthAverage number of live streaming classes watchedAverage number of live streaming classes broadcasted
930 84689 04371 960 7727780875
Table 1 Statistics of live streaming dataset.
UserItemInteractionAverage length per userAverage length per item
59 60038 201262 79346
Table 2 Statistics of the Amazon dataset.
DatasetMethodPerformance
PrecisionRecallF1NDCG
HuajiaoDARM0.1860.4660.2660.392
ADARM0.2400.6000.3420.541
AmazonDARM0.1920.5240.2790.403
ADARM0.2490.6890.3630.672
Table 3 Performance of DARM and ADARM for top 5 recommendations.
DatasetMethodPerformance
PrecisionRecallF1NDCG
HuajiaoDARM0.1260.6330.2110.478
ADARM0.1360.6830.2270.574
AmazonDARM0.1250.6780.2100.464
ADARM0.1350.7450.2270.693
Table 4 Performance of DARM and ADARM for top 10 recommendations.
DatasetMethodPerformance
PrecisionRecallF1NDCG
HuajiaoMF0.0800.2010.1150.186
BPR0.1040.2600.1490.243
STAMP0.1420.3560.2030.305
SHAN0.1530.3830.2180.323
SR-GNN0.1660.4160.2380.369
TiSASRec0.1730.4330.2470.372
RRLC0.2060.5160.2950.458
ADARM (Improvement)0.240 (+16.50%)0.600 (+16.27%)0.342 (+15.93%)0.541 (+18.12%)
AmazonMF0.0450.1130.0640.083
BPR0.0850.2130.1220.169
STAMP0.1240.3350.1790.275
SHAN0.1480.4020.2150.301
SR-GNN0.1620.4430.2350.339
TiSASRec0.1690.4600.2450.359
RRLC0.1970.5350.2860.526
ADARM (Improvement)0.249 (+26.39%)0.689 (+28.78%)0.363 (+26.92%)0.672 (+27.75%)
Table 5 Performance comparison among different models for top 5 recommendations.
DatasetMethodPerformance
PrecisionRecallF1NDCG
HuajiaoMF0.0470.2350.0780.199
BPR0.0580.2940.0980.257
STAMP0.0820.4130.1370.328
SHAN0.0940.4730.1570.359
SR-GNN0.1160.5830.1940.446
TiSASRec0.1230.6160.2050.459
RRLC0.1330.6660.2220.517
ADARM (Improvement)0.136 (+2.25%)0.683 (+2.55%)0.227 (+2.25%)0.574 (+11.02%)
AmazonMF0.0440.2190.0730.125
BPR0.0590.2950.0980.202
STAMP0.0850.4700.1440.326
SHAN0.1020.5560.1710.313
SR-GNN0.1080.5850.1810.395
TiSASRec0.1210.6600.2040.436
RRLC0.1320.7230.2230.598
ADARM (Improvement)0.135 (+2.27%)0.745 (+3.04%)0.227 (+1.79%)0.693 (+15.88%)
Table 6 Performance comparison among different models for top 10 recommendations.
Fig. 4 Recall with different numbers of heads on the Huajiao dataset.
Fig. 5 NDCG with different numbers of heads on the Huajiao dataset.
[1]   Chen X., Nguyen T. V., Shen Z. Q., and Kankanhalli M., Livesense: Contextual advertising in live streaming videos, in Proc. 27th ACM Int. Conf. on Multimedia, Nice, France, 2019, pp. 392-400.
[2]   Zhu Z. H., Yang Z., and Dai Y. F., Understanding the gift-sending interaction on live-streaming video websites, in Proc. 9th Int. Conf. on Social Computing and Social Media, Vancouver, Canada, 2017, pp. 274-285.
[3]   Lu Z. C., Xia H. J., Heo S., and Wigdor D., You watch, you give, and you engage: A study of live streaming practices in China, in Proc. 2018 CHI Conf. on Human Factors in Computing Systems, Montreal, Canada, 2018, pp. 1-13.
[4]   Koren Y., Bell R., and Volinsky C., Matrix factorization techniques for recommender systems, Computer, vol. 42, no. 8, pp. 30-37, 2009.
[5]   He R. N. and McAuley J., Fusing similarity models with markov chains for sparse sequential recommendation, in Proc. 2016 IEEE 16th Int. Conf. on Data Mining, Barcelona, Spain, 2016, pp. 191-200.
[6]   Yu F., Liu Q., Wu S., Wang L., and Tan T. N., A dynamic recurrent model for next basket recommendation, in Proc. 39th Int. ACM SIGIR Conf. on Research and Development in Information Retrieval, Pisa, Italy, 2016, pp. 729-732.
[7]   Hidasi B., Karatzoglou A., Baltrunas L., and Tikk D., Session-based recommendations with recurrent neural networks, in Proc. 4th Int. Conf. on Learning Representations, San Juan, Puerto Rico, 2016.
[8]   Tan Y. K., Xu X. X., and Liu Y., Improved recurrent neural networks for session-based recommendations, in Proc. 1st Workshop on Deep Learning for Recommender Systems, Boston, MA, USA, 2016, pp. 17-22.
[9]   Wu C. Y., Ahmed A., Beutel A., Smola A. J., and Jing H., Recurrent recommender networks, in Proc. 10th ACM Int. Conf. on Web Search and Data Mining, Cambridge, UK, 2017, pp. 495-503.
[10]   Li Z., Zhao H. K., Liu Q., Huang Z. Y., Mei T., and Chen E. H., Learning from history and present: Next-item recommendation via discriminatively exploiting user behaviors, in Proc. 24th ACM SIGKDD Int. Conf. on Knowledge Discovery & Data Mining, London, UK, 2018, pp. 1734-1743.
[11]   Chen X., Xu H. T., Zhang Y. F., Tang J. X., Cao Y. X., Qin Z., and Zha H. Y., Sequential recommendation with user memory networks, in Proc. 11th ACM Int. Conf. on Web Search and Data Mining, Marina Del Rey, CA, USA, 2018, pp. 108-116.
[12]   Huang J., Zhao W. X., Dou H. J., Wen J. R., and Chang E. Y., Improving sequential recommendation with knowledge-enhanced memory networks, in Proc. 41st Int. ACM SIGIR Conf. on Research & Development in Information Retrieval, Ann Arbor, MI, USA, 2018, pp. 505-514.
[13]   Zhou M. Z., Ding Z. Y., Tang J. L., and Yin D. W., Micro behaviors: A new perspective in e-commerce recommender systems, in Proc. 11th ACM Int. Conf. on Web Search and Data Mining, Marina Del Rey, CA, USA, 2018, pp. 727-735.
[14]   Ying H. C., Zhuang F. Z., Zhang F. Z., Liu Y. C., Xu G. D., Xie X., Xiong H., and Wu J., Sequential recommender system based on hierarchical attention network, in Proc. 27th Int. Joint Conf. on Artificial Intelligence, Stockholm, Sweden, 2018, pp. 3926-3932.
[15]   Liu Q., Zeng Y. F., Mokhosi R., and Zhang H. B., Stamp: Short-term attention/memory priority model for session-based recommendation, in Proc. 24th ACM SIGKDD Int. Conf. on Knowledge Discovery & Data Mining, London, UK, 2018, pp. 1831-1839.
[16]   Yu L., Zhang C. X., Liang S. S., and Zhang X. L., Multi-order attentive ranking model for sequential recommendation, in Proc. 33rd AAAI Conf. on Artificial Intelligence, Honolulu, HI, USA, 2019, pp. 5709-5716.
[17]   Wu S., Tang Y. Y., Zhu Y. Q., Wang L., Xie X., and Tan T. J., Session-based recommendation with graph neural networks, in Proc. 33rd AAAI Conf. on Artificial Intelligence, Honolulu, HI, USA, 2019, pp. 346-353.
[18]   Zhang T. T., Zhao P. P., Liu Y. C., Sheng V. S., Xu J. J., Wang D. Q., Liu G. F., and Zhou X. F., Feature-level deeper self-attention network for sequential recommendation, in Proc. 28th Int. Joint Conf. on Artificial Intelligence, Macao, China, 2019, pp. 4320-4326.
[19]   Wang J. L. and Caverlee J., Recurrent recommendation with local coherence, in Proc. 12th ACM Int. Conf. on Web Search and Data Mining, Melbourne, Australia, 2019, pp. 564-572.
[20]   Li J. C., Wang Y. J., and McAuley J., Time interval aware self-attention for sequential recommendation, in Proc. 13th Int. Conf. on Web Search and Data Mining, Houston, TX, USA, 2020, pp. 322-330.
[21]   Mikolov T., Chen K., Corrado G., and Dean J., Efficient estimation of word representations in vector space, in Proc. 1st Int. Conf. on Learning Representations, Scottsdale, AZ, USA, 2013.
[22]   Hochreiter S., Schmidhuber J., Long short-term memory, Neural Computation, vol. 9, no. 8, pp. 1735-1780, 1997.
[23]   Rendle S., Freudenthaler C., Gantner Z., and Schmidt-Thieme L., BPR: Bayesian personalized ranking from implicit feedback, in Proc. 25th Conf. on Uncertainty in Artificial Intelligence, Montreal, Canada, 2009, pp. 452-461.
[24]   Tang J. X. and Wang K., Personalized top-n sequential recommendation via convolutional sequence embedding, in Proc. 11th ACM Int. Conf. on Web Search and Data Mining, Marina Del Rey, CA, USA, 2018, pp. 565-573.
[1] Natarajan Yuvaraj,Kannan Srihari,Selvaraj Chandragandhi,Rajan Arshath Raja,Gaurav Dhiman,Amandeep Kaur. Analysis of Protein-Ligand Interactions of SARS-CoV-2 Against Selective Drug Using Deep Neural Networks[J]. Big Data Mining and Analytics, 2021, 4(2): 76-83.
[2] Krishna Kant Singh,Akansha Singh. Diagnosis of COVID-19 from Chest X-Ray Images Using Wavelets-Based Depthwise Convolution Network[J]. Big Data Mining and Analytics, 2021, 4(2): 84-93.
[3] Youssef Nait Malek,Mehdi Najib,Mohamed Bakhouya,Mohammed Essaaidi. Multivariate Deep Learning Approach for Electric Vehicle Speed Forecasting[J]. Big Data Mining and Analytics, 2021, 4(1): 56-64.
[4] Wei Zhong, Ning Yu, Chunyu Ai. Applying Big Data Based Deep Learning System to Intrusion Detection[J]. Big Data Mining and Analytics, 2020, 3(3): 181-195.
[5] Sunitha Basodi, Chunyan Ji, Haiping Zhang, Yi Pan. Gradient Amplification: An Efficient Way to Train Deep Neural Networks[J]. Big Data Mining and Analytics, 2020, 3(3): 196-207.
[6] Chaity Banerjee, Tathagata Mukherjee, Eduardo Pasiliao Jr.. Feature Representations Using the Reflected Rectified Linear Unit (RReLU) Activation[J]. Big Data Mining and Analytics, 2020, 3(2): 102-120.
[7] Zhenxing Guo, Shihua Zhang. Sparse Deep Nonnegative Matrix Factorization[J]. Big Data Mining and Analytics, 2020, 03(01): 13-28.
[8] Yang Yang, Nengjun Zhu, Yifeng Wu, Jian Cao, Dechuan Zhan, Hui Xiong. A Semi-Supervised Attention Model for Identifying Authentic Sneakers[J]. Big Data Mining and Analytics, 2020, 03(01): 29-40.
[9] Qile Zhu, Xiyao Ma, Xiaolin Li. Statistical Learning for Semantic Parsing: A Survey[J]. Big Data Mining and Analytics, 2019, 2(4): 217-239.
[10] Ying Yu, Min Li, Liangliang Liu, Yaohang Li, Jianxin Wang. Clinical Big Data and Deep Learning: Applications, Challenges, and Future Outlooks[J]. Big Data Mining and Analytics, 2019, 2(4): 288-305.
[11] Wenmao Wu, Zhizhou Yu, Jieyue He. A Semi-Supervised Deep Network Embedding Approach Based on the Neighborhood Structure[J]. Big Data Mining and Analytics, 2019, 2(3): 205-216.
[12] Jiangcheng Zhu, Shuang Hu, Rossella Arcucci, Chao Xu, Jihong Zhu, Yi-ke Guo. Model Error Correction in Data Assimilation by Integrating Neural Networks[J]. Big Data Mining and Analytics, 2019, 2(2): 83-91.
[13] Jin Liu, Yi Pan, Min Li, Ziyue Chen, Lu Tang, Chengqian Lu, Jianxin Wang. Applications of Deep Learning to MRI Images: A Survey[J]. Big Data Mining and Analytics, 2018, 1(1): 1-18.
[14] Ning Yu, Zhihua Li, Zeng Yu. Survey on Encoding Schemes for Genomic Data Representation and Feature Learning—From Signal Processing to Machine Learning[J]. Big Data Mining and Analytics, 2018, 01(03): 191-210.
[15] Qianyu Meng, Kun Wang, Xiaoming He, Minyi Guo. QoE-Driven Big Data Management in Pervasive Edge Computing Environment[J]. Big Data Mining and Analytics, 2018, 01(03): 222-233.