Advertising platforms have a growing need for improving prediction quality, as missing out on ad opportunities can have a negative effect on their performance. To that end, prediction tasks such as conversion prediction need to be continuously advanced through the inclusion of data from new sources or through algorithmic development that tackles existing challenges. The introduction of different data sources naturally brings unwanted noise, whereas underexplored areas still exist in modeling approaches, such as temporal information of events in sequences. In this study, we propose extensions for modeling online user activity trails that address two very important aspects of activities—time and noise, through dedicated layers that can be used in existing deep sequence-learning approaches. Our proposed method exhibited area under the receiver operating characteristic curve improvement of up to 3% and 1.75% compared with production and best baseline approaches, respectively, across two major advertiser data sets and several predictive tasks.
Get full access to this article
View all access options for this article.
References
1.
GligorijevicD, ZhouT, BharatbhushanS, et al.Bid shading in the brave new world of first-price auctions. In: d'Aquin M, Dietze A, Hauff C, et al. (Eds.): Proceedings of the 29th ACM International Conference on Information and Knowledge Management, CIKM 2020, Virtual Event, Ireland, October 19–23, 2020. pp. 2453–2460.
2.
HeX, PanJ, JinO, et al.Practical lessons from predicting clicks on ads at Facebook. In: Saka E, Shed D, Lee K, et al. (Eds.): Proceedings of the 8th International Workshop on Data Mining for Online Advertising, AdKDD 2014. ACM, New York, NY, August 24–27, 2014, pp. 1–9.
3.
GligorijevicD, StojanovicJ, RaghuveerA, et al.Modeling mobile user actions for purchase recommendations using deep memory networks. In: Collins-Thompson K, Mei Q, Davison B, et al. (Eds.): Proceedings of the 41st International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR. 2018. pp. 1021–1024.
4.
ZhouY, MishraS, GligorijevicJ, et al.Understanding consumer journey using attention based recurrent neural networks. In: Teredesai A, Kumar V, Li Y, et al. (Eds.): Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, KDD 2019, Anchorage, AK, August 4–8, 2019. pp. 3102–3111.
5.
BahdanauD, ChoK, BengioY. Neural machine translation by jointly learning to align and translate. In: Bengio Y, LeCun Y, Kingsbury B, et al. (Eds.): Proceedings of the 3rd International Conference on Learning Representations, ICLR 2015, San Diego, CA, May 7–9, 2015.
6.
KarlssonN.Control problems in online advertising and benefits of randomized bidding strategies. Eur J Control. 2016; 30:31–49.
7.
BhamidipatiN, KantR, MishraS.A large scale prediction engine for app install clicks and conversions. In: Lim E, Winslett M, Sanderson M, et al. (Eds.): Proceedings of the 26th International Conference on Information and Knowledge Management, CIKM 2017, Singapore, November 6–10, 2017. pp. 167–175.
8.
RichardsonM, DominowskaE, RagnoR.Predicting clicks: Estimating the click-through rate for new ads. In: Williamson C, Zurko ME, Patel-Scheider P, et al. (Eds.): Proceedings of the 16th International Conference on World Wide Web, WWW 2007, Banff, Alberta, Canada, May 8–12, 2007, pp. 521–530.
9.
McMahanHB, HoltG, SculleyD, et al.Ad click prediction: A view from the trenches. In: Ghani R, Senator T, Bradley P, et al. (Eds.): Proceedings of the 19th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2013, Chicago, IL, August 11–14, 2013. pp. 1222–1230.
10.
PanJ, MaoY, RuizAL, et al.Predicting different types of conversions with multi-task learning in online advertising. In: Teredesai A, Kumar V, Li Y, et al. (Eds.): Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, KDD 2019, Anchorage, AK, August 4–8, 2019. pp. 1834–1842.
11.
ShanY, HoensTR, JiaoJ, et al.Deep crossing: Web-scale modeling without manually crafted combinatorial features. In: Krishnapuram B, Shah M, Aggrawal C, et al. (Eds.): Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2016, San Francisco, CL, August 13–17, 2016. pp. 255–262.
12.
GligorijevicJ, GligorijevicD, StojkovicI, et al.Deeply supervised model for click-through rate prediction in sponsored search. Data Min Knowl Disc. 2019; 33:1446–1467.
13.
JiangZ.Research on ctr prediction for contextual advertising based on deep architecture model. J Control Eng Appl Inform. 2016; 180:11–19.
14.
ZhangY, DaiH, XuC, et al.Sequential click prediction for sponsored search with recurrent neural networks. In: Brodley C, Stone P (Eds.): Twenty-Eighth AAAI Conference on Artificial Intelligence, Quebec City, Quebec, Canada, July 27–31, 2014.
15.
CuiY, TobossiR, VigourouxO. Modelling customer online behaviours with neural networks: Applications to conversion prediction and advertising retargeting. arXiv preprint arXiv:1804.07669, 2018.
16.
AravaSK, DongC, YanZ, et al.Deep neural net with attention for multichannel multi-touch attribution. arXiv preprint arXiv:1809.02230, 2018.
17.
PeiW, TaxDMJ. Unsupervised learning of sequence representations by autoencoders. arXiv preprint arXiv:1804.00946, 2018.
18.
BeutelA, CovingtonP, JainS, et al.Latent cross: Making use of context in recurrent recommender systems. In: Chang Y, Zhai C, Liu Y, et al. (Eds.): Proceedings of the Eleventh ACM International Conference on Web Search and Data Mining, WSDM 2018. ACM, Marina Del Rey, CA, February 5–9, 2018. pp. 46–54.
19.
LiY, DuN, BengioS.Time-dependent representation for neural event sequence prediction. In: Bengio Y, LeCun Y, Sainath T, et al. (Eds.): 6th International Conference on Learning Representations, ICLR 2018, Vancouver, BC, Canada, April 30–May 3, 2018, Workshop Track Proceedings.
20.
JingH, SmolaAJ. Neural survival recommender. In: de Rijke M, Shokouhi M, Tomkins A, et al. (Eds.): Proceedings of the Tenth ACM International Conference on Web Search and Data Mining, WSDM 2017, Cambridge, United Kingdom, February 6–10, 2017. pp. 515–524.
21.
ZhuY, LiH, LiaoY, et al.What to do next: Modeling user behaviors by time-LSTM. In: Sierra C (Eds.): Proceedings of the 26th International Joint Conference on Artificial Intelligence, IJCAI 2017, Melbourne, Australia, August 19–25, 2017. pp. 3602–3608.
22.
RajkomarA, OrenE, ChenK, et al.Scalable and accurate deep learning with electronic health records. NPJ Digital Med. 2018; 1:18.
23.
GligorijevicD, GligorijevicJ, FloresA.Prospective modeling of users for online display advertising via deep time-aware model. In: d'Aquin M, Dietze A, Hauff C, et al. (Eds.): Proceedings of the 29th ACM International Conference on Information and Knowledge Management, CIKM 2020, Virtual Event, Ireland, October 19–23, 2020. pp. 2461–2468.
24.
SutskeverI, VinyalsO, LeQV. Sequence to sequence learning with neural networks. In: Ghahramani Z, Welling M, Cortes C, et al. (Eds.): Proceedings of the 27th International Conference on Neural Information Processing Systems, NeurIPS 2014, Montreal, Canada, December 8–13 2014. pp. 3104–3112.
25.
VaswaniA, ShazeerN, ParmarN, et al.Attention is all you need. In: von Luxburg U, Guyon I, Bengio S, et al. (Eds.): Proceedings of the 31st International Conference on Neural Information Processing Systems, NeurIPS 2017, Long Beach, CA, December 4–9, 2017. pp. 5998–6008.
26.
GligorijevicD, StojanovicJ, SatzW, et al.Deep attention model for triage of emergency department patients. In: Obradovic Z, Parthasarathy S, Apte C, et al. (Eds.): Proceedings of the 2018 SIAM International Conference on Data Mining, SDM 2018, San Diego, CA, May 3–5, 2018. pp. 297–305.
27.
XuK, BaJ, KirosR, et al.Show, attend and tell: Neural image caption generation with visual attention. In: Bach F, Blei D (Eds.): Proceedings of the 32nd International Conference on International Conference on Machine Learning, ICML 2015, Lille, France, July 6–11, 2015. pp. 2048–2057.
28.
MetzL, IbarzJ, JaitlyN, DavidsonJ. Discrete sequential prediction of continuous actions for deep RL. arXiv preprint arXiv:1705.05035, 2017.
29.
Kaiser q, BengioS.Discrete autoencoders for sequence models. arXiv preprint arXiv:1801.09797, 2018.
30.
BengioY, LeonardN, CourvilleA. Estimating or propagating gradients through stochastic neurons for conditional computation. arXiv preprint arXiv:1308.3432, 2013.
MaddisonCJ, MnihA, TehYW. The concrete distribution: A continuous relaxation of discrete random variables. In: Bengio Y, LeCum Y, Ranzato M, et al. (Eds.): Proceedings of 5th International Conference on Learning Representations, ICLR 2017, Toulon, France, April 24–26, 2017.
34.
JangE, GuS, PooleB. Categorical reparameterization with Gumbel-Softmax. In: Bengio Y, LeCum Y, Ranzato M, et al. (Eds.): Proceedings of 5th International Conference on Learning Representations, ICLR 2017, Toulon, France, April 24–26, 2017.
35.
SalakhutdinovR, HintonG. Semantic hashing. Int J Approx Reason. 2009; 50:969–978.
36.
ShenC, QiG-J, JiangR, et al.Sharp attention network via adaptive sampling for person re-identification. IEEE Trans Circuits Syst Video Technol. 2019; 29:3016–3027.
37.
YanS, SmithJS, LuW, ZhangB. Hierarchical multi-scale attention networks for action recognition. Signal Process Image Commun. 2018; 61:73–84.
38.
KaiserL, BengioS.Can active memory replace attention? In: Lee D, von Luxburg U, Garnett R, et al. (Eds.) Proceedings of the 30th International Conference on Neural Information Processing Systems, NeurIPS 2016, Barcelona, Spain, December 5–10, 2016. pp. 3781–3789.
39.
LouizosC, WellingM, KingmaDP. Learning sparse neural networks through l 0 regularization. In: Bengio Y, LeCun Y, Sainath T, et al. (Eds.): 6th International Conference on Learning Representations, ICLR 2018, Vancouver, BC, Canada, April 30–May 3, 2018.
40.
CaoXH, HanC, ObradovicZ. Learning a dynamic-based representation for multivariate biomarker time series classifications. In: 2018IEEE International Conference on Healthcare Informatics, ICHI 2018, IEEE, 2018. pp. 163–173.
41.
BaiT, ZhangS, EglestonBL, VuceticS.Interpretable representation learning for healthcare via capturing disease progression through time. In: Guo Y, Farooq (eds.): Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, KDD 2018, London, United Kingdom, August 19–23, 2018. pp. 43–51.
42.
SchusterM, PaliwalKK. Bidirectional recurrent neural networks. IEEE Trans Signal Process. 1997; 450:2673–2681.
43.
ChoK, van MerrienboerB, GulcehreC, et al.Learning phrase representations using RNN encoder–decoder for statistical machine translation. In: Moschitti A, Pang B, Alfonseca E, et al. (Eds.): Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing, EMNLP 2014, Doha, Qatar, October 25–29, 2014. pp. 1724–1734.
44.
ZhaiS, ChangK-H, ZhangR, ZhangZM. Deepintent: Learning attentions for online advertising with recurrent neural networks. In: Krishnapuram B, Shah M, Aggrawal C, et al. (Eds.): Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2016, San Francisco, California, August 13–17, 2016. pp. 1295–1304.
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
