Deep Interest Network for Taobao advertising data Click-Through Rate Prediction

Di Sun

doi:10.1109/CISCE52179.2021.9445990

Deep Interest Network for Taobao advertising data Click-Through Rate Prediction

Di Sun^*

^*Corresponding author for this work

Department of Computer Science

Research output: Chapter in book/report/conference proceeding › Conference proceeding › peer-review

2 Citations (Scopus)

Abstract

Click-through rate prediction has been widely used in recommendation system, search engine optimization, advertising and other fields. Traditional machine learning methods such as LR, FM, and GBDT require a large amount of artificial feature construction processes and only learn the linear relationship between features. With the rapid development of deep learning in natural language processing, computer vision, many studies such as DeepFM, PNN, and WideDeep, apply DNN models to Click-through rate prediction estimation. In this work, I use the modified deep interest network (DIN) for the click-through rate prediction task. The mixed loss is proposed as the training loss to fit the DIN model. Attention mechanism is improved according to different input features. I evaluate proposed deep interest network model on Taobao ad click data. Results show the model has superior performance than other deep learning methods. Experiments show that the modified DIN model is effective for the Taobao advertising data click-through rate prediction task.

Original language	English
Title of host publication	Proceedings of 2021 IEEE 3rd International Conference on Communications, Information System and Computer Engineering, CISCE 2021
Publisher	IEEE
Pages	741-744
Number of pages	4
ISBN (Electronic)	9780738112152, 9780738112145
DOIs	https://doi.org/10.1109/CISCE52179.2021.9445990
Publication status	Published - 14 May 2021
Event	3rd IEEE International Conference on Communications, Information System and Computer Engineering, CISCE 2021 - Beijing, China Duration: 14 May 2021 → 16 May 2021 https://ieeexplore.ieee.org/xpl/conhome/9445845/proceeding

Publication series

Name	Proceedings of International Conference on Communications, Information System and Computer Engineering (CISCE)
Publisher	IEEE

Conference

Conference	3rd IEEE International Conference on Communications, Information System and Computer Engineering, CISCE 2021
Country/Territory	China
City	Beijing
Period	14/05/21 → 16/05/21
Internet address	https://ieeexplore.ieee.org/xpl/conhome/9445845/proceeding

User-Defined Keywords

Click-through Rate Prediction
Deep Interest Network
Deep Neural Network
Recommender Systems

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1109/CISCE52179.2021.9445990

Cite this

Sun, D. (2021). Deep Interest Network for Taobao advertising data Click-Through Rate Prediction. In Proceedings of 2021 IEEE 3rd International Conference on Communications, Information System and Computer Engineering, CISCE 2021 (pp. 741-744). (Proceedings of International Conference on Communications, Information System and Computer Engineering (CISCE)). IEEE. https://doi.org/10.1109/CISCE52179.2021.9445990

@inproceedings{187ad57f4c104b52bdb2fdc4cfefbfa8,

title = "Deep Interest Network for Taobao advertising data Click-Through Rate Prediction",

abstract = "Click-through rate prediction has been widely used in recommendation system, search engine optimization, advertising and other fields. Traditional machine learning methods such as LR, FM, and GBDT require a large amount of artificial feature construction processes and only learn the linear relationship between features. With the rapid development of deep learning in natural language processing, computer vision, many studies such as DeepFM, PNN, and WideDeep, apply DNN models to Click-through rate prediction estimation. In this work, I use the modified deep interest network (DIN) for the click-through rate prediction task. The mixed loss is proposed as the training loss to fit the DIN model. Attention mechanism is improved according to different input features. I evaluate proposed deep interest network model on Taobao ad click data. Results show the model has superior performance than other deep learning methods. Experiments show that the modified DIN model is effective for the Taobao advertising data click-through rate prediction task.",

keywords = "Click-through Rate Prediction, Deep Interest Network, Deep Neural Network, Recommender Systems",

author = "Di Sun",

note = "Publisher Copyright: {\textcopyright} 2021 IEEE.; 3rd IEEE International Conference on Communications, Information System and Computer Engineering, CISCE 2021 ; Conference date: 14-05-2021 Through 16-05-2021",

year = "2021",

month = may,

day = "14",

doi = "10.1109/CISCE52179.2021.9445990",

language = "English",

series = "Proceedings of International Conference on Communications, Information System and Computer Engineering (CISCE)",

publisher = "IEEE",

pages = "741--744",

booktitle = "Proceedings of 2021 IEEE 3rd International Conference on Communications, Information System and Computer Engineering, CISCE 2021",

address = "United States",

url = "https://ieeexplore.ieee.org/xpl/conhome/9445845/proceeding",

}

Sun, D 2021, Deep Interest Network for Taobao advertising data Click-Through Rate Prediction. in Proceedings of 2021 IEEE 3rd International Conference on Communications, Information System and Computer Engineering, CISCE 2021. Proceedings of International Conference on Communications, Information System and Computer Engineering (CISCE), IEEE, pp. 741-744, 3rd IEEE International Conference on Communications, Information System and Computer Engineering, CISCE 2021, Beijing, China, 14/05/21. https://doi.org/10.1109/CISCE52179.2021.9445990

Deep Interest Network for Taobao advertising data Click-Through Rate Prediction. / Sun, Di.
Proceedings of 2021 IEEE 3rd International Conference on Communications, Information System and Computer Engineering, CISCE 2021. IEEE, 2021. p. 741-744 (Proceedings of International Conference on Communications, Information System and Computer Engineering (CISCE)).

Research output: Chapter in book/report/conference proceeding › Conference proceeding › peer-review

TY - GEN

T1 - Deep Interest Network for Taobao advertising data Click-Through Rate Prediction

AU - Sun, Di

PY - 2021/5/14

Y1 - 2021/5/14

N2 - Click-through rate prediction has been widely used in recommendation system, search engine optimization, advertising and other fields. Traditional machine learning methods such as LR, FM, and GBDT require a large amount of artificial feature construction processes and only learn the linear relationship between features. With the rapid development of deep learning in natural language processing, computer vision, many studies such as DeepFM, PNN, and WideDeep, apply DNN models to Click-through rate prediction estimation. In this work, I use the modified deep interest network (DIN) for the click-through rate prediction task. The mixed loss is proposed as the training loss to fit the DIN model. Attention mechanism is improved according to different input features. I evaluate proposed deep interest network model on Taobao ad click data. Results show the model has superior performance than other deep learning methods. Experiments show that the modified DIN model is effective for the Taobao advertising data click-through rate prediction task.

AB - Click-through rate prediction has been widely used in recommendation system, search engine optimization, advertising and other fields. Traditional machine learning methods such as LR, FM, and GBDT require a large amount of artificial feature construction processes and only learn the linear relationship between features. With the rapid development of deep learning in natural language processing, computer vision, many studies such as DeepFM, PNN, and WideDeep, apply DNN models to Click-through rate prediction estimation. In this work, I use the modified deep interest network (DIN) for the click-through rate prediction task. The mixed loss is proposed as the training loss to fit the DIN model. Attention mechanism is improved according to different input features. I evaluate proposed deep interest network model on Taobao ad click data. Results show the model has superior performance than other deep learning methods. Experiments show that the modified DIN model is effective for the Taobao advertising data click-through rate prediction task.

KW - Click-through Rate Prediction

KW - Deep Interest Network

KW - Deep Neural Network

KW - Recommender Systems

UR - http://www.scopus.com/inward/record.url?scp=85111439562&partnerID=8YFLogxK

U2 - 10.1109/CISCE52179.2021.9445990

DO - 10.1109/CISCE52179.2021.9445990

M3 - Conference proceeding

AN - SCOPUS:85111439562

T3 - Proceedings of International Conference on Communications, Information System and Computer Engineering (CISCE)

SP - 741

EP - 744

BT - Proceedings of 2021 IEEE 3rd International Conference on Communications, Information System and Computer Engineering, CISCE 2021

PB - IEEE

T2 - 3rd IEEE International Conference on Communications, Information System and Computer Engineering, CISCE 2021

Y2 - 14 May 2021 through 16 May 2021

ER -

Sun D. Deep Interest Network for Taobao advertising data Click-Through Rate Prediction. In Proceedings of 2021 IEEE 3rd International Conference on Communications, Information System and Computer Engineering, CISCE 2021. IEEE. 2021. p. 741-744. (Proceedings of International Conference on Communications, Information System and Computer Engineering (CISCE)). doi: 10.1109/CISCE52179.2021.9445990

Deep Interest Network for Taobao advertising data Click-Through Rate Prediction

Abstract

Publication series

Conference

User-Defined Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this