Abstract

User behavior modeling is essential in computational advertisement, which builds users' profiles by tracking their online behaviors and then delivers the relevant ads according to each user's interests and needs. Accurate models will lead to higher targeting accuracy and thus improved advertising performance. Intuitively, similar users tend to have similar behaviors towards the displayed ads (e.g., impression, click, conversion). However, to the best of our knowledge, there is not much previous work that explicitly investigates such similarities of various types of user behaviors, and incorporates them into ad response targeting and prediction, largely due to the prohibitive scale of the problem. To bridge this gap, in this paper, we use bipartite graphs to represent historical user behaviors, which consist of both user nodes and advertiser campaign nodes, as well as edges reflecting various types of user-campaign interactions in the past. Based on this representation, we study random-walk-based local algorithms for user behavior modeling and action prediction, whose computational complexity depends only on the size of the output cluster, rather than the entire graph. Our goal is to improve action prediction by leveraging historical user-user, campaign-campaign, and user-campaign interactions. In particular, we propose the bipartite graphs AdvUserGraph accompanied with the ADNI algorithm. ADNI extends the NIBBLE algorithm to AdvUserGraph, and it is able tofi nd the local cluster consisting of interested users towards a specific advertiser campaign. We also propose two extensions of ADNI with improved efficiencies. The performance of the proposed algorithms is demonstrated on both synthetic data and a world leading Demand Side Platform (DSP), showing that they are able to discriminate extremely rare events in terms of their action propensity.

Original languageEnglish (US)
Title of host publicationKDD 2017 - Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
PublisherAssociation for Computing Machinery
Pages2091-2099
Number of pages9
VolumePart F129685
ISBN (Electronic)9781450348874
DOIs
StatePublished - Aug 13 2017
Event23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2017 - Halifax, Canada
Duration: Aug 13 2017Aug 17 2017

Other

Other23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2017
CountryCanada
CityHalifax
Period8/13/178/17/17

Fingerprint

Display devices
Marketing
Computational complexity

Keywords

  • Computational advertisement
  • Large scale
  • Local graph algorithm
  • User action prediction

ASJC Scopus subject areas

  • Software
  • Information Systems

Cite this

Yang, H., Zhu, Y., & He, J. (2017). Local algorithm for user action prediction towards display ads. In KDD 2017 - Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (Vol. Part F129685, pp. 2091-2099). Association for Computing Machinery. https://doi.org/10.1145/3097983.3098089

Local algorithm for user action prediction towards display ads. / Yang, Hongxia; Zhu, Yada; He, Jingrui.

KDD 2017 - Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Vol. Part F129685 Association for Computing Machinery, 2017. p. 2091-2099.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Yang, H, Zhu, Y & He, J 2017, Local algorithm for user action prediction towards display ads. in KDD 2017 - Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. vol. Part F129685, Association for Computing Machinery, pp. 2091-2099, 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2017, Halifax, Canada, 8/13/17. https://doi.org/10.1145/3097983.3098089
Yang H, Zhu Y, He J. Local algorithm for user action prediction towards display ads. In KDD 2017 - Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Vol. Part F129685. Association for Computing Machinery. 2017. p. 2091-2099 https://doi.org/10.1145/3097983.3098089
Yang, Hongxia ; Zhu, Yada ; He, Jingrui. / Local algorithm for user action prediction towards display ads. KDD 2017 - Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Vol. Part F129685 Association for Computing Machinery, 2017. pp. 2091-2099
@inproceedings{355c928343d84c82be6b7528b35d61c4,
title = "Local algorithm for user action prediction towards display ads",
abstract = "User behavior modeling is essential in computational advertisement, which builds users' profiles by tracking their online behaviors and then delivers the relevant ads according to each user's interests and needs. Accurate models will lead to higher targeting accuracy and thus improved advertising performance. Intuitively, similar users tend to have similar behaviors towards the displayed ads (e.g., impression, click, conversion). However, to the best of our knowledge, there is not much previous work that explicitly investigates such similarities of various types of user behaviors, and incorporates them into ad response targeting and prediction, largely due to the prohibitive scale of the problem. To bridge this gap, in this paper, we use bipartite graphs to represent historical user behaviors, which consist of both user nodes and advertiser campaign nodes, as well as edges reflecting various types of user-campaign interactions in the past. Based on this representation, we study random-walk-based local algorithms for user behavior modeling and action prediction, whose computational complexity depends only on the size of the output cluster, rather than the entire graph. Our goal is to improve action prediction by leveraging historical user-user, campaign-campaign, and user-campaign interactions. In particular, we propose the bipartite graphs AdvUserGraph accompanied with the ADNI algorithm. ADNI extends the NIBBLE algorithm to AdvUserGraph, and it is able tofi nd the local cluster consisting of interested users towards a specific advertiser campaign. We also propose two extensions of ADNI with improved efficiencies. The performance of the proposed algorithms is demonstrated on both synthetic data and a world leading Demand Side Platform (DSP), showing that they are able to discriminate extremely rare events in terms of their action propensity.",
keywords = "Computational advertisement, Large scale, Local graph algorithm, User action prediction",
author = "Hongxia Yang and Yada Zhu and Jingrui He",
year = "2017",
month = "8",
day = "13",
doi = "10.1145/3097983.3098089",
language = "English (US)",
volume = "Part F129685",
pages = "2091--2099",
booktitle = "KDD 2017 - Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining",
publisher = "Association for Computing Machinery",

}

TY - GEN

T1 - Local algorithm for user action prediction towards display ads

AU - Yang, Hongxia

AU - Zhu, Yada

AU - He, Jingrui

PY - 2017/8/13

Y1 - 2017/8/13

N2 - User behavior modeling is essential in computational advertisement, which builds users' profiles by tracking their online behaviors and then delivers the relevant ads according to each user's interests and needs. Accurate models will lead to higher targeting accuracy and thus improved advertising performance. Intuitively, similar users tend to have similar behaviors towards the displayed ads (e.g., impression, click, conversion). However, to the best of our knowledge, there is not much previous work that explicitly investigates such similarities of various types of user behaviors, and incorporates them into ad response targeting and prediction, largely due to the prohibitive scale of the problem. To bridge this gap, in this paper, we use bipartite graphs to represent historical user behaviors, which consist of both user nodes and advertiser campaign nodes, as well as edges reflecting various types of user-campaign interactions in the past. Based on this representation, we study random-walk-based local algorithms for user behavior modeling and action prediction, whose computational complexity depends only on the size of the output cluster, rather than the entire graph. Our goal is to improve action prediction by leveraging historical user-user, campaign-campaign, and user-campaign interactions. In particular, we propose the bipartite graphs AdvUserGraph accompanied with the ADNI algorithm. ADNI extends the NIBBLE algorithm to AdvUserGraph, and it is able tofi nd the local cluster consisting of interested users towards a specific advertiser campaign. We also propose two extensions of ADNI with improved efficiencies. The performance of the proposed algorithms is demonstrated on both synthetic data and a world leading Demand Side Platform (DSP), showing that they are able to discriminate extremely rare events in terms of their action propensity.

AB - User behavior modeling is essential in computational advertisement, which builds users' profiles by tracking their online behaviors and then delivers the relevant ads according to each user's interests and needs. Accurate models will lead to higher targeting accuracy and thus improved advertising performance. Intuitively, similar users tend to have similar behaviors towards the displayed ads (e.g., impression, click, conversion). However, to the best of our knowledge, there is not much previous work that explicitly investigates such similarities of various types of user behaviors, and incorporates them into ad response targeting and prediction, largely due to the prohibitive scale of the problem. To bridge this gap, in this paper, we use bipartite graphs to represent historical user behaviors, which consist of both user nodes and advertiser campaign nodes, as well as edges reflecting various types of user-campaign interactions in the past. Based on this representation, we study random-walk-based local algorithms for user behavior modeling and action prediction, whose computational complexity depends only on the size of the output cluster, rather than the entire graph. Our goal is to improve action prediction by leveraging historical user-user, campaign-campaign, and user-campaign interactions. In particular, we propose the bipartite graphs AdvUserGraph accompanied with the ADNI algorithm. ADNI extends the NIBBLE algorithm to AdvUserGraph, and it is able tofi nd the local cluster consisting of interested users towards a specific advertiser campaign. We also propose two extensions of ADNI with improved efficiencies. The performance of the proposed algorithms is demonstrated on both synthetic data and a world leading Demand Side Platform (DSP), showing that they are able to discriminate extremely rare events in terms of their action propensity.

KW - Computational advertisement

KW - Large scale

KW - Local graph algorithm

KW - User action prediction

UR - http://www.scopus.com/inward/record.url?scp=85029024066&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85029024066&partnerID=8YFLogxK

U2 - 10.1145/3097983.3098089

DO - 10.1145/3097983.3098089

M3 - Conference contribution

VL - Part F129685

SP - 2091

EP - 2099

BT - KDD 2017 - Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

PB - Association for Computing Machinery

ER -