請用此 Handle URI 來引用此文件：
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/70784完整後設資料紀錄
| DC 欄位 | 值 | 語言 |
|---|---|---|
| dc.contributor.advisor | 盧信銘(Hsin-Min Lu) | |
| dc.contributor.author | Yao-Yu Tsai | en |
| dc.contributor.author | 蔡曜宇 | zh_TW |
| dc.date.accessioned | 2021-06-17T04:38:22Z | - |
| dc.date.available | 2019-08-08 | |
| dc.date.copyright | 2018-08-08 | |
| dc.date.issued | 2018 | |
| dc.date.submitted | 2018-08-07 | |
| dc.identifier.citation | 參考文獻
Avila, C. P., & Vijaya, M. (2016). Click through rate prediction for display advertisement. International Journal of Computer Applications, 136(1). Broder, A. Z. (2008). Computational advertising. Paper presented at the SODA. Chapelle, O. (2014). Modeling delayed feedback in display advertising. Paper presented at the Proceedings of the 20th ACM SIGKDD international conference on Knowledge discovery and data mining. Chapelle, O., Manavoglu, E., & Rosales, R. (2015). Simple and scalable response prediction for display advertising. ACM Transactions on Intelligent Systems and Technology (TIST), 5(4), 61. Chen, J.-H., Zhao, Z.-Q., Shi, J.-Y., & Zhao, C. (2017). A New Approach for Mobile Advertising Click-Through Rate Estimation Based on Deep Belief Nets. Computational intelligence and neuroscience, 2017. Chen, J., Sun, B., Li, H., Lu, H., & Hua, X.-S. (2016a). Deep ctr prediction in display advertising. Paper presented at the Proceedings of the 2016 ACM on Multimedia Conference. Chen, Q.-H., Yu, S.-M., Guo, Z.-X., & Jia, Y.-B. (2016b). Estimating Ads’ Click through Rate with Recurrent Neural Network. Paper presented at the ITM Web of Conferences. Cheng, H.-T., Koc, L., Harmsen, J., Shaked, T., Chandra, T., Aradhye, H., . . . Ispir, M. (2016). Wide & deep learning for recommender systems. Paper presented at the Proceedings of the 1st Workshop on Deep Learning for Recommender Systems. Cheng, H., Zwol, R. v., Azimi, J., Manavoglu, E., Zhang, R., Zhou, Y., & Navalpakkam, V. (2012). Multimedia features for click prediction of new ads in display advertising. Paper presented at the Proceedings of the 18th ACM SIGKDD international conference on Knowledge discovery and data mining. Chung, J., Gulcehre, C., Cho, K., & Bengio, Y. (2014). Empirical evaluation of gated recurrent neural networks on sequence modeling. Presented in NIPS 2014 Deep Learning and Representation Learning Workshop. Cui, Y., Zhang, R., Li, W., & Mao, J. (2011). Bid landscape forecasting in online ad exchange marketplace. Paper presented at the Proceedings of the 17th ACM SIGKDD international conference on Knowledge discovery and data mining. Dave, K. S., & Varma, V. (2010). Learning the click-through rate for rare/new ads from similar ads. Paper presented at the Proceedings of the 33rd international ACM SIGIR conference on Research and development in information retrieval. Dembczynski, K., Kotlowski, W., & Weiss, D. (2008). Predicting ads clickthrough rate with decision rules. Paper presented at the Workshop on targeting and ranking in online advertising. Friedman, J. H. (2001). Greedy function approximation: a gradient boosting machine. Annals of statistics, 1189-1232. Ghosh, A., Rubinstein, B. I., Vassilvitskii, S., & Zinkevich, M. (2009). Adaptive bidding for display advertising. Paper presented at the Proceedings of the 18th international conference on World wide web. Graves, A., Mohamed, A.-r., & Hinton, G. (2013). Speech recognition with deep recurrent neural networks. Paper presented at the Acoustics, speech and signal processing (icassp), 2013 ieee international conference on. Guo, H., Tang, R., Ye, Y., Li, Z., & He, X. (2017). DeepFM: A Factorization-Machine based Neural Network for CTR Prediction. In IJCAI. 2782–2788. He, X., Pan, J., Jin, O., Xu, T., Liu, B., Xu, T., . . . Bowers, S. (2014). Practical lessons from predicting clicks on ads at facebook. Paper presented at the Proceedings of the Eighth International Workshop on Data Mining for Online Advertising. Hinton, G. E., & Salakhutdinov, R. R. (2006). Reducing the dimensionality of data with neural networks. science, 313(5786), 504-507. Hochreiter, S., & Schmidhuber, J. (1997). Long short-term memory. Neural computation, 9(8), 1735-1780. Kondakindi, G., Rana, S., Rajkumar, A., Ponnekanti, S. K., & Parakh, V. (2014). A logistic regression approach to ad click prediction. Mach. Learn.-Cl. Proj. Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2012). Imagenet classification with deep convolutional neural networks. Paper presented at the Advances in neural information processing systems. Lang, T., & Rettenmeier, M. (2017). Understanding consumer behavior with recurrent neural networks. Paper presented at the International Workshop on Machine Learning Methods for Recommender Systems. LeCun, Y., Bengio, Y., & Hinton, G. (2015). Deep learning. nature, 521(7553), 436. Lee, K.-C., Jalali, A., & Dasdan, A. (2013). Real time bid optimization with smooth budget delivery in online advertising. Paper presented at the Proceedings of the Seventh International Workshop on Data Mining for Online Advertising. Lee, K.-c., Orten, B., Dasdan, A., & Li, W. (2012). Estimating conversion rate in display advertising from past erformance data. Paper presented at the Proceedings of the 18th ACM SIGKDD international conference on Knowledge discovery and data mining. Lipton, Z. C., Berkowitz, J., & Elkan, C. (2015). A critical review of recurrent neural networks for sequence learning. CoRR, 1506.00019, 2015. Liu, Q., Yu, F., Wu, S., & Wang, L. (2015). A convolutional click prediction model. Paper presented at the Proceedings of the 24th ACM International on Conference on Information and Knowledge Management. McMahan, H. B., Holt, G., Sculley, D., Young, M., Ebner, D., Grady, J., . . . Golovin, D. (2013). Ad click prediction: a view from the trenches. Paper presented at the Proceedings of the 19th ACM SIGKDD international conference on Knowledge discovery and data mining. Pan, Z., Chen, E., Liu, Q., Xu, T., Ma, H., & Lin, H. (2016). Sparse Factorization Machines for Click-through Rate Prediction. Paper presented at the Data Mining (ICDM), 2016 IEEE 16th International Conference on. Ren, K., Zhang, W., Rong, Y., Zhang, H., Yu, Y., & Wang, J. (2016). User response learning for directly optimizing campaign performance in display advertising. Paper presented at the Proceedings of the 25th ACM International on Conference on Information and Knowledge Management. Rendle, S. (2012a). Factorization machines with libfm. ACM Transactions on Intelligent Systems and Technology (TIST), 3(3), 57. Rendle, S. (2012b). Social network and click-through prediction with factorization machines. Paper presented at the KDD-Cup Workshop. Richardson, M., Dominowska, E., & Ragno, R. (2007). Predicting clicks: estimating the click-through rate for new ads. Paper presented at the Proceedings of the 16th international conference on World Wide Web. Shen, S., Hu, B., Chen, W., & Yang, Q. (2012). Personalized click model through collaborative filtering. Paper presented at the Proceedings of the fifth ACM international conference on Web search and data mining. Ta, A.-P. (2015). Factorization machines with follow-the-regularized-leader for CTR prediction in display advertising. Paper presented at the Big Data (Big Data), 2015 IEEE International Conference on. Trofimov, I., Kornetova, A., & Topinskiy, V. (2012). Using boosted trees for clickthrough rate prediction for sponsored search. Paper presented at the Proceedings of the Sixth International Workshop on Data Mining for Online Advertising and Internet Economy. Vincent, P., Larochelle, H., Bengio, Y., & Manzagol, P.-A. (2008). Extracting and composing robust features with denoising autoencoders. Paper presented at the Proceedings of the 25th international conference on Machine learning. Wang, F., Suphamitmongkol, W., & Wang, B. (2013). Advertisement click-through rate prediction using multiple criteria linear programming regression model. Procedia Computer Science, 17, 803-811. Wang, R., Fu, B., Fu, G., & Wang, M. (2017). Deep & Cross Network for Ad Click Predictions. In Proceedings of the ADKDD'17 (p. 12). ACM. Wu, W. C.-H., Yeh, M.-Y., & Chen, M.-S. (2015). Predicting winning price in real time bidding with censored data. Paper presented at the Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Yan, L., Li, W.-j., Xue, G.-R., & Han, D. (2014). Coupled group lasso for web-scale ctr prediction in display advertising. Paper presented at the International Conference on Machine Learning. Yi, J., Chen, Y., Li, J., Sett, S., & Yan, T. W. (2013). Predictive model performance: Offline and online evaluations. Paper presented at the Proceedings of the 19th ACM SIGKDD international conference on Knowledge discovery and data mining. Yuan, S., Wang, J., & Zhao, X. (2013). Real-time bidding for online advertising: measurement and analysis. Paper presented at the Proceedings of the Seventh International Workshop on Data Mining for Online Advertising. Yuan, Y., Wang, F., Li, J., & Qin, R. (2014). A survey on real time bidding advertising. Paper presented at the Service Operations and Logistics, and Informatics (SOLI), 2014 IEEE International Conference on. Zhang, W., Du, T., & Wang, J. (2016). Deep learning over multi-field categorical data. Paper presented at the European conference on information retrieval. Zhang, W., Yuan, S., Wang, J., & Shen, X. (2014a). Real-time bidding benchmarking with iPinYou dataset. Bid 501.015300008 (2014): 3f5a4f5121. Zhang, W., Zhang, Y., Gao, B., Yu, Y., Yuan, X., & Liu, T.-Y. (2012). Joint optimization of bid and budget allocation in sponsored search. Paper presented at the Proceedings of the 18th ACM SIGKDD international conference on Knowledge discovery and data mining. Zhang, Y., Dai, H., Xu, C., Feng, J., Wang, T., Bian, J., . . . Liu, T.-Y. (2014b). Sequential Click Prediction for Sponsored Search with Recurrent Neural Networks. Paper presented at the AAAI. Zhou, G., Song, C., Zhu, X., Ma, X., Yan, Y., Dai, X., . . . Gai, K. (2017). Deep Interest Network for Click-Through Rate Prediction. stat 1050 (2017): 23. | |
| dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/70784 | - |
| dc.description.abstract | 廣告點擊率預測是線上廣告的重要研究方向之一，隨著即時競價機制興起與公開資料公布，廣告研究重心從贊助搜索廣告轉移到即時競價中的展示型廣告，大部分研究運用機器學習預測使用者在假設每筆廣告曝光為獨立的情況是否點擊，但使用者真實的網站點擊行為是會受到先前使用經驗影響。所以本篇研究除了將文獻提出的廣告點擊率預測模型應用於展示型廣告資料集上，更以使用者為中心參考使用者歷史廣告特徵的序列實驗方式來設計出不同的廣告點擊假設，探討新的實驗假設是否能提升模型對廣告點擊率預測的表現。
本篇研究使用的是 Avazu 公司在 Kaggle 平台舉辦的廣告點擊率預測競賽資料集，將此次實驗分別以三個實驗假設進行設計與實做探討，分別為 (1) 單一獨立廣告實驗假設 (2) 考慮使用者歷史廣告展示序列實驗假設 (針對時間相關特徵) (3) 考慮使用者歷史廣告展示序列實驗假設 (考慮所有相關特徵)。接著從預測模型與特徵工程方面提出模型改良與特徵工程來提升點擊率預測表現，我們提出名為 CNN&GRU 模型，主要運用 Wide&Deep 模型架構來結合卷積神經網路與 GRU (Gated Recurrent Unit) 模型，針對使用者歷史點擊經驗會影響下次點擊率的特性，將歷史點擊經驗量化為新特徵進行預測。最後探討序列實驗的長度變化與類神經網路模型架構對模型在各實驗假設預測表現的影響性。 實驗結果反映出模型結構不同會導致模型學習特徵方式與特徵資訊也會不同，並影響模型在不同實驗假設的表現差異。接著，在只運用時間特徵資訊的序列假設下無法有效進行點擊率預測，但在運用歷史廣告相關特徵的序列假設下能有效幫助模型提高預測表現。實驗結果也證實所提出的 CNN&GRU 模型與特徵工程能有效提升點廣告擊率預測表現，並且歷史特徵序列長度增加與類神經相關模型的內部構造參數調整也會提升模型對廣告點擊率預測表現。 | zh_TW |
| dc.description.abstract | Advertising click through rate prediction is one of the fundamental problems in online advertising. Researchers have already adopted machine learning approaches to predict advertising click for each ad view independently. However, as observed user’s behaviors on ads yield high dependency on how the user behaved along with the past time. Therefore, in addition to applying the models to the display advertising dataset, we also
design ad click hypotheses based on user historical advertising features. In this study, we propose a model called CNN&GRU that uses the Wide&Deep model architecture to combine the convolutional neural network with the GRU (Gated Recurrent Unit) model, and then discusses the influence of feature engineering, sequence experiment length variation and neural network related model architecture. In addition, we used Avazu's advertising click-through rate forecasting contest dataset on the Kaggle platform to examine our proposed model. The experiment was implemented with independent and sequence hypotheses. Our results show that different model structures will affect the performance in different hypotheses. Using only the time feature information can not effectively predict the click-through rate, but the sequence hypothesis can effectively improve the forecast performance. In addition, CNN&GRU model, feature engineering and the increase of the historical feature sequence can also improve the prediction performance. | en |
| dc.description.provenance | Made available in DSpace on 2021-06-17T04:38:22Z (GMT). No. of bitstreams: 1 ntu-107-R05725012-1.pdf: 3200583 bytes, checksum: 97685e4a4fbe3552c581dcff87c0d6ea (MD5) Previous issue date: 2018 | en |
| dc.description.tableofcontents | 主目錄
口試委員審定書………………………………………………………………………………….i 誌謝………………………………………………………………………………………………ii 摘要……………………………………………………………………………………………...iii Abstract ....……………………………………………………………………………………….iv 主目錄....…………………………………………………………………………………………v 圖目錄………………………………………………………………………………………….viii 表目錄……………………………………………………………………………………………x 第一章 緒論 ............................................................................................................................. 1 1.1 研究背景與動機..................................................................................................... 1 1.2 研究目的................................................................................................................. 5 1.3 研究架構................................................................................................................. 5 第二章 文獻探討......................................................................................................................... 6 2.1 即時競價的投標策略與得標價預測..................................................................... 6 2.1.1 競標價格策略......................................................................................................... 7 2.1.2 廣告點擊率重要性................................................................................................. 9 2.2 廣告的點擊與轉換率預測................................................................................... 11 2.2.1 廣告點擊率特徵處理........................................................................................... 14 2.2.2 廣告點擊率預測模型 (線性模型) ...................................................................... 16 2.2.3 廣告點擊率預測模型 (非線性模型)................................................................... 18 2.2.4 廣告點擊率預測模型 (類神經相關模型)........................................................... 21 2.2.5 廣告點擊率預測模型 (混合相關模型)............................................................... 27 2.2.6 廣告點擊率在線學習........................................................................................... 31 2.2.7 廣告轉換率預測................................................................................................... 33 2.3 序列性廣告研究與模型....................................................................................... 34 2.4 小結....................................................................................................................... 37 第三章 研究方法 ................................................................................................................... 38 3.1 資料來源及處理................................................................................................... 38 3.1.1 資料前處理........................................................................................................... 39 3.2 研究流程............................................................................................................... 44 3.2.1 單一獨立廣告實驗假設....................................................................................... 45 3.2.2 考慮使用者歷史廣告展示序列實驗假設(針對時間相關特徵)......................... 46 3.2.3 考慮使用者歷史廣告展示序列實驗假設(考慮所有相關特徵)......................... 48 3.3 預測模型............................................................................................................... 50 3.3.1 邏輯迴歸模型....................................................................................................... 51 3.3.2 梯度提升決策樹模型........................................................................................... 52 3.3.3 因式分解機模型................................................................................................... 53 3.3.4 類神經網路模型................................................................................................... 54 3.3.5 卷積神經網路模型............................................................................................... 55 3.3.6 循環神經網路相關模型....................................................................................... 56 3.3.7 梯度提升決策樹模型 + 邏輯迴歸模型............................................................. 57 3.3.8 Wide & Deep 模型............................................................................................... 58 3.4 特徵工程與模型設計........................................................................................... 60 3.4.1 特徵工程............................................................................................................... 60 3.4.2 模型設計............................................................................................................... 61 3.5 序列長度與類神經模型架構影響....................................................................... 63 3.6 衡量指標............................................................................................................... 65 3.6.1 AUC 衡量指標...................................................................................................... 65 3.6.2 Logloss 衡量指標 ................................................................................................. 66 第四章 結果與討論 ............................................................................................................... 67 4.1 實驗假設結果....................................................................................................... 67 4.1.1 實驗假設一結果................................................................................................... 67 4.1.2 實驗假設二結果................................................................................................... 69 4.1.3 實驗假設三結果................................................................................................... 71 4.1.4 實驗假設比較....................................................................................................... 73 4.2 模型設計與特徵工程........................................................................................... 75 4.2.1 模型設計結果....................................................................................................... 75 4.2.2 特徵工程結果....................................................................................................... 77 4.3 序列長度影響性探討........................................................................................... 79 4.3.1. 實驗假設二與假設三結果................................................................................... 79 4.4 類神經相關模型架構探討................................................................................... 81 4.4.1 類神經網路模型架構探討................................................................................... 81 4.4.2 卷積神經網路模型架構探討............................................................................... 83 第五章 結論與建議 ............................................................................................................... 85 5.1 研究結果............................................................................................................... 85 5.2 研究貢獻............................................................................................................... 86 5.3 未來方向............................................................................................................... 87 參考文獻......................................................................................................................................88 | |
| dc.language.iso | zh-TW | |
| dc.subject | 序列性廣告點擊 | zh_TW |
| dc.subject | 廣告點擊率 | zh_TW |
| dc.subject | 推薦系統 | zh_TW |
| dc.subject | 展示型廣告 | zh_TW |
| dc.subject | 混合模型 | zh_TW |
| dc.subject | Hybrid Models | en |
| dc.subject | Recommder System | en |
| dc.subject | Display Ads | en |
| dc.subject | Ad Click-Through Rate | en |
| dc.subject | Sequence Ads Clicks | en |
| dc.title | 展示型廣告點擊率預測模型 : 比較與應用 | zh_TW |
| dc.title | Click Through Rate Predict for Display Advertising :
Comparisons and Applications | en |
| dc.type | Thesis | |
| dc.date.schoolyear | 106-2 | |
| dc.description.degree | 碩士 | |
| dc.contributor.oralexamcommittee | 洪為璽(Wei-Hsi Hung),余峻瑜(Jun-Yu Yu) | |
| dc.subject.keyword | 廣告點擊率,推薦系統,展示型廣告,序列性廣告點擊,混合模型, | zh_TW |
| dc.subject.keyword | Ad Click-Through Rate,Recommder System,Display Ads,Sequence Ads Clicks,Hybrid Models, | en |
| dc.relation.page | 91 | |
| dc.identifier.doi | 10.6342/NTU201802540 | |
| dc.rights.note | 有償授權 | |
| dc.date.accepted | 2018-08-08 | |
| dc.contributor.author-college | 管理學院 | zh_TW |
| dc.contributor.author-dept | 資訊管理學研究所 | zh_TW |
| 顯示於系所單位： | 資訊管理學系 | |
文件中的檔案：
| 檔案 | 大小 | 格式 | |
|---|---|---|---|
| ntu-107-1.pdf 未授權公開取用 | 3.13 MB | Adobe PDF |
系統中的文件，除了特別指名其著作權條款之外，均受到著作權保護，並且保留所有的權利。