1. NCCU Academic Hub
  2. Publications
  3. College of Social Science
  4. Theses
  5. Item

Publications-Theses

Article View/Open

Publication Export

Google ScholarTM

NCCU Library

Citation Infomation

Related Publications in TAIR

題名 BeCross：融合個人行為偏誤的推薦系統建模方法
BeCross: A Personalized Behavior Bias Modeling Layer for Deep Recommender Systems
作者 楊宗泰
Yang, Tsung-Tai
貢獻者 徐士勛<br>蔡炎龍
Hsu, Shih-Hsun<br>Tsai, Yen-lung
楊宗泰
Yang, Tsung-Tai
關鍵詞 推薦系統
行為經濟學交互推薦系統
行為經濟學
個人化
Recommendation system
BeCrossRec
Behavior economics
Personalized
日期 2025
上傳時間 4-Aug-2025 12:48:48 (UTC+8)
摘要 推薦系統自引入深度學習以來持續蓬勃發展。Neural Collaborative Filtering （NCF） 將協同過濾嵌入深度架構以學習非線性交互；Deep Crossing 促成特徵自動組合；Wide & Deep 結合記憶與泛化能力；DIN 引入注意力機制用以學習用戶歷史行為與目標項目的關聯；MaskNet 則修正了 DNN 對於乘法交互表達力不足的問題。這些研究皆回應了實務上推薦系統面臨的挑戰，並透過架構設計不斷提升模型能力。推薦系統本質是一種協助人類決策的工具，過去研究多未考慮用戶決策本身所蘊含的偏誤與捷思。根據行為經濟學理論，人在面對決策選擇時常受錨定效應、可得性捷思等影響，例如用戶觀看 Netflix 的漫威影集時，其期待值會受到過往觀看經驗的影響，並在此基礎上進行調整。然而若推薦模型未能捕捉這類心理偏誤，將導致重要變數遺漏與預測偏差。為此，本文提出 BeCrossRec 架構，在現有深度學習推薦架構中，嵌入行為經濟學特徵交互模組，並透過特徵化、個人化捷思與偏誤，使模型得以捕捉多種個人化行為偏誤之複雜規則，並於實驗結果發現有效提升預測準確度。此設計不僅納入人性決策捷思與偏誤，也為未來結合行為經濟學理論與推薦系統研究，提供可行之方向。
Since the introduction of deep learning, recommender systems have undergone rapid and sustained advancement. Neural Collaborative Filtering integrates collaborative filtering into deep architectures to capture nonlinear interactions; Deep Crossing enables automatic feature combination; Wide & Deep networks balance memorization and generalization capabilities; Deep Interest Network introduces attention mechanisms to model the relationship between user history and target items; and MaskNet addresses the limitations of DNNs in expressing multiplicative interactions. These approaches have tackled real-world challenges in recommendation tasks and continuously improved algorithmic performance through architectural innovations. Fundamentally, recommender systems serve as tools to assist human decision-making. However, most existing studies have overlooked the cognitive biases and heuristics that influence user decisions. According to behavioral economics, individuals are often affected by phenomena such as the anchoring effect and availability heuristic when making choices. For example, a user's expectations while watching a Marvel series on Netflix may be shaped by their prior viewing experiences, which serve as anchors for current judgments. If recommendation models fail to capture such psychological biases, they risk omitting critical variables and introducing prediction bias. To address this, we propose BeCrossRec, a novel architecture that integrates a behavioral economics feature interaction module into existing deep recommendation frameworks. By explicitly modeling heuristics and biases—both in general and in personalized forms—BeCrossRec enables the learning of complex rules underlying individual behavioral biases. Experimental results demonstrate that this approach significantly improves predictive accuracy. This design not only incorporates human decision-making tendencies into recommendation modeling but also offers a viable direction for future research at the intersection of behavioral economics and recommender systems.
參考文獻 [1]. Adomavicius, G., Bockstedt, J. C., Curley, S. P., & Zhang, J. (2013). Do recommender systems manipulate consumer preferences? A study of anchoring effects. Information Systems Research, 24(4), 956-975. [2]. Anil, R., Gadanho, S., Huang, D., Jacob, N., Li, Z., Lin, D., ... & Yan, Q. (2022). On the factory floor: ML engineering for industrial-scale ads recommendation models. arXiv preprint arXiv:2209.05310. [3]. Chen, J., Dong, H., Wang, X., Feng, F., Wang, M., & He, X. (2023). Bias and debias in recommender system: A survey and future directions. ACM Transactions on Information Systems, 41(3), 1-39. [4]. Chen, Q., Zhao, H., Li, W., Huang, P., & Ou, W. (2019, August). Behavior sequence transformer for e-commerce recommendation in alibaba. In Proceedings of the 1st international workshop on deep learning practice for high-dimensional sparse data (pp. 1-4). [5]. Cheng, H. T., Koc, L., Harmsen, J., Shaked, T., Chandra, T., Aradhye, H., ... & Shah, H. (2016, September). Wide & deep learning for recommender systems. In Proceedings of the 1st workshop on deep learning for recommender systems (pp. 7-10). [6]. Chiny, M., Chihab, M., Bencharef, O., & Chihab, Y. (2022). Netflix recommendation system based on TF-IDF and cosine similarity algorithms. no. Bml, 15-20. [7]. Clarke, K. A. (2005). The phantom menace: Omitted variable bias in econometric research. Conflict management and peace science, 22(4), 341-352. [8]. Covington, P., Adams, J., & Sargin, E. (2016, September). Deep neural networks for youtube recommendations. In Proceedings of the 10th ACM conference on recommender systems (pp. 191-198). [9]. Kingma, D. P., & Ba, J. (2015). Adam: A method for stochastic optimization. In Proceedings of the 3rd International Conference on Learning Representations (ICLR). [10]. Ge, Y., Xu, S., Liu, S., Fu, Z., Sun, F., & Zhang, Y. (2020, July). Learning personalized risk preferences for recommendation. In Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval (pp. 409-418). [11]. Harper, F. M., & Konstan, J. A. (2015). The movielens datasets: History and context. Acm transactions on interactive intelligent systems (tiis), 5(4), 1-19. [12]. He, X., Liao, L., Zhang, H., Nie, L., Hu, X., & Chua, T. S. (2017, April). Neural collaborative filtering. In Proceedings of the 26th international conference on world wide web (pp. 173-182). [13]. Hornik, K. (1991). Approximation capabilities of multilayer feedforward networks. Neural networks, 4(2), 251-257. [14]. Houlsby, N., Giurgiu, A., Jastrzebski, S., Morrone, B., De Laroussilhe, Q., Gesmundo, A., ... & Gelly, S. (2019, May). Parameter-efficient transfer learning for NLP. In International conference on machine learning (pp. 2790-2799). PMLR. [15]. Hrnjic, E., & Tomczak, N. (2019). Machine learning and behavioral economics for personalized choice architecture. arXiv preprint arXiv:1907.02100. [16]. Jesse, M., & Jannach, D. (2021). Digital nudging with recommender systems: Survey and future directions. Computers in Human Behavior Reports, 3, 100052. [17]. Kahneman, D., & Tversky, A. (2013). Prospect theory: An analysis of decision under risk. In Handbook of the fundamentals of financial decision making: Part I (pp. 99-127). [18]. Koren, Y., Bell, R., & Volinsky, C. (2009). Matrix factorization techniques for recommender systems. Computer, 42(8), 30-37. [19]. Lex, E., Kowald, D., Seitlinger, P., Tran, T. N. T., Felfernig, A., & Schedl, M. (2021). Psychology-informed recommender systems. Foundations and trends® in information retrieval, 15(2), 134-242. [20]. Naumov, M., Mudigere, D., Shi, H. J. M., Huang, J., Sundaraman, N., Park, J., ... & Smelyanskiy, M. (2019). Deep learning recommendation model for personalization and recommendation systems. arXiv preprint arXiv:1906.00091. [21]. Raza, S., Rahman, M., Kamawal, S., Toroghi, A., Raval, A., Navah, F., & Kazemeini, A. (2024). A comprehensive review of recommender systems: Transitioning from theory to practice. arXiv preprint arXiv:2407.13699. [22]. Soto, C. J., & Jackson, J. J. (2013). Five-factor model of personality. Journal of Research in Personality, 42, 1285-1302. [23]. Sparck Jones, K. (1972). A statistical interpretation of term specificity and its application in retrieval. Journal of documentation, 28(1), 11-21. [24]. Steck, H., Baltrunas, L., Elahi, E., Liang, D., Raimond, Y., & Basilico, J. (2021). Deep learning for recommender systems: A Netflix case study. AI magazine, 42(3), 7-18. [25]. Sun, F., Liu, J., Wu, J., Pei, C., Lin, X., Ou, W., & Jiang, P. (2019, November). BERT4Rec: Sequential recommendation with bidirectional encoder representations from transformer. In Proceedings of the 28th ACM international conference on information and knowledge management (pp. 1441-1450). [26]. Thaler, R. H. (2016). Behavioral economics: Past, present, and future. American economic review, 106(7), 1577-1600. [27]. Thaler, R. H., & Sunstein, C. R. (2009). Nudge: Improving decisions about health, wealth, and happiness. Penguin. [28]. Tkalcic, M., Kosir, A., & Tasic, J. (2011). Affective recommender systems: the role of emotions in recommender systems. In The RecSys 2011 Workshops-Decisions@ RecSys 2011 and UCERSTI-2: Human Decision Making in Recommender Systems; User-Centric Evaluation of Recommender Systems and Their Interfaces-2 (Vol. 811, pp. 9-13). CEUR-WS. org. [29]. Tversky, A., & Kahneman, D. (1973). Availability: A heuristic for judging frequency and probability. Cognitive psychology, 5(2), 207-232. [30]. Tversky, A., & Kahneman, D. (1974). Judgment under Uncertainty: Heuristics and Biases: Biases in judgments reveal some heuristics of thinking under uncertainty. science, 185(4157), 1124-1131. [31]. Tversky, A., & Kahneman, D. (1981). The framing of decisions and the psychology of choice. science, 211(4481), 453-458. [32]. Uber Engineering. (2022, March 23). Graph learning powering the Uber Eats graph ecosystem. Uber Blog. https://www.uber.com/en-TW/blog/uber-eats-graph-learning/ [33]. Wang, R., Fu, B., Fu, G., & Wang, M. (2017). Deep & cross network for ad click predictions. In Proceedings of the ADKDD'17 (pp. 1-7). [34]. Wang, R., Shivanna, R., Cheng, D., Jain, S., Lin, D., Hong, L., & Chi, E. (2021, April). Dcn v2: Improved deep & cross network and practical lessons for web-scale learning to rank systems. In Proceedings of the web conference 2021 (pp. 1785-1797). [35]. Wang, T., Brovman, Y. M., & Madhvanath, S. (2021). Personalized embedding-based e-commerce recommendations at ebay. arXiv preprint arXiv:2102.06156. [36]. Wang, Z., She, Q., & Zhang, J. (2021). Masknet: Introducing feature-wise multiplication to CTR ranking models by instance-guided mask. arXiv preprint arXiv:2102.07619. [37]. Xia, X., Eksombatchai, P., Pancha, N., Badani, D. D., Wang, P. W., Gu, N., ... & Zhai, A. (2023, August). Transact: Transformer-based realtime user action model for recommendation at pinterest. In Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (pp. 5249-5259). [38]. Xiao, Y., Zhang, Y., & Li, X. (2024, June). modeling variational anchoring effect for recommender systems. In 2024 IEEE Conference on Artificial Intelligence (CAI) (pp. 926-931). IEEE. [39]. Zhao, Z., Hong, L., Wei, L., Chen, J., Nath, A., Andrews, S., ... & Chi, E. (2019, September). Recommending what video to watch next: a multitask ranking system. In Proceedings of the 13th ACM conference on recommender systems (pp. 43-51). [40]. Zhou, G., Zhu, X., Song, C., Fan, Y., Zhu, H., Ma, X., ... & Gai, K. (2018, July). Deep interest network for click-through rate prediction. In Proceedings of the 24th ACM SIGKDD international conference on knowledge discovery & data mining (pp. 1059-1068).
描述 碩士
國立政治大學
經濟學系
108258037
資料來源 http://thesis.lib.nccu.edu.tw/record/#G0108258037
資料類型 thesis
dc.contributor.advisor 徐士勛<br>蔡炎龍zh_TW
dc.contributor.advisor Hsu, Shih-Hsun<br>Tsai, Yen-lungen_US
dc.contributor.author (Authors) 楊宗泰zh_TW
dc.contributor.author (Authors) Yang, Tsung-Taien_US
dc.creator (作者) 楊宗泰zh_TW
dc.creator (作者) Yang, Tsung-Taien_US
dc.date (日期) 2025en_US
dc.date.accessioned 4-Aug-2025 12:48:48 (UTC+8)-
dc.date.available 4-Aug-2025 12:48:48 (UTC+8)-
dc.date.issued (上傳時間) 4-Aug-2025 12:48:48 (UTC+8)-
dc.identifier (Other Identifiers) G0108258037en_US
dc.identifier.uri (URI) https://nccur.lib.nccu.edu.tw/handle/140.119/158265-
dc.description (描述) 碩士zh_TW
dc.description (描述) 國立政治大學zh_TW
dc.description (描述) 經濟學系zh_TW
dc.description (描述) 108258037zh_TW
dc.description.abstract (摘要) 推薦系統自引入深度學習以來持續蓬勃發展。Neural Collaborative Filtering （NCF） 將協同過濾嵌入深度架構以學習非線性交互；Deep Crossing 促成特徵自動組合；Wide & Deep 結合記憶與泛化能力；DIN 引入注意力機制用以學習用戶歷史行為與目標項目的關聯；MaskNet 則修正了 DNN 對於乘法交互表達力不足的問題。這些研究皆回應了實務上推薦系統面臨的挑戰，並透過架構設計不斷提升模型能力。推薦系統本質是一種協助人類決策的工具，過去研究多未考慮用戶決策本身所蘊含的偏誤與捷思。根據行為經濟學理論，人在面對決策選擇時常受錨定效應、可得性捷思等影響，例如用戶觀看 Netflix 的漫威影集時，其期待值會受到過往觀看經驗的影響，並在此基礎上進行調整。然而若推薦模型未能捕捉這類心理偏誤，將導致重要變數遺漏與預測偏差。為此，本文提出 BeCrossRec 架構，在現有深度學習推薦架構中，嵌入行為經濟學特徵交互模組，並透過特徵化、個人化捷思與偏誤，使模型得以捕捉多種個人化行為偏誤之複雜規則，並於實驗結果發現有效提升預測準確度。此設計不僅納入人性決策捷思與偏誤，也為未來結合行為經濟學理論與推薦系統研究，提供可行之方向。zh_TW
dc.description.abstract (摘要) Since the introduction of deep learning, recommender systems have undergone rapid and sustained advancement. Neural Collaborative Filtering integrates collaborative filtering into deep architectures to capture nonlinear interactions; Deep Crossing enables automatic feature combination; Wide & Deep networks balance memorization and generalization capabilities; Deep Interest Network introduces attention mechanisms to model the relationship between user history and target items; and MaskNet addresses the limitations of DNNs in expressing multiplicative interactions. These approaches have tackled real-world challenges in recommendation tasks and continuously improved algorithmic performance through architectural innovations. Fundamentally, recommender systems serve as tools to assist human decision-making. However, most existing studies have overlooked the cognitive biases and heuristics that influence user decisions. According to behavioral economics, individuals are often affected by phenomena such as the anchoring effect and availability heuristic when making choices. For example, a user's expectations while watching a Marvel series on Netflix may be shaped by their prior viewing experiences, which serve as anchors for current judgments. If recommendation models fail to capture such psychological biases, they risk omitting critical variables and introducing prediction bias. To address this, we propose BeCrossRec, a novel architecture that integrates a behavioral economics feature interaction module into existing deep recommendation frameworks. By explicitly modeling heuristics and biases—both in general and in personalized forms—BeCrossRec enables the learning of complex rules underlying individual behavioral biases. Experimental results demonstrate that this approach significantly improves predictive accuracy. This design not only incorporates human decision-making tendencies into recommendation modeling but also offers a viable direction for future research at the intersection of behavioral economics and recommender systems.en_US
dc.description.tableofcontents 摘要 2 Abstract 3 目錄 4 表次 6 圖次 7 第一章 緒論 8 1.1 前言 8 1.2 研究目的 9 第二章 文獻回顧 10 2.1 推薦系統介紹 10 2.1.1推薦系統的分類 11 2.1.2 推薦系統如何消除偏誤 12 2.2 行為經濟學理論基礎 13 2.2.1 起源 13 2.2.2 捷思與偏誤 14 2.3 行為經濟學與推薦系統的結合 17 2.3.1 心理學與推薦系統 17 2.3.2 行為經濟學與推薦系統 18 2.4 研究空白與方向 19 第三章 理論依據與主要方法 20 3.1 問題定義 20 3.1.1 特徵化捷思與偏誤 20 3.1.2 個人化捷思與偏誤 21 3.1.3 多種捷思與偏誤之交互影響探討 22 3.2 BeCrossRec 架構 23 第四章 資料與實證結果分析 26 4.1 資料集 26 4.1.1 資料集簡介 26 4.1.2 資料處理 27 4.1.3 特徵工程 27 4.1.4 資料切分 28 4.2 實驗設計 28 4.2.1 評估指標 28 4.2.2 基準模型說明 29 4.2.3 實驗設定 30 4.3 實驗結果 30 4.3.1 Q1：特徵化捷思與偏誤是否有效提升模型預測力？ 30 4.3.2 Q2：個人化捷思與偏誤是否有效提升模型預測力？ 31 4.3.3 Q3：多種捷思與偏誤交互是否有效提升模型預測力？ 31 4.3.4 Q4：BeCrossRec 是否有效提升模型預測力？ 32 第五章 結論 34 參考文獻 35zh_TW
dc.format.extent 1328327 bytes-
dc.format.mimetype application/pdf-
dc.source.uri (資料來源) http://thesis.lib.nccu.edu.tw/record/#G0108258037en_US
dc.subject (關鍵詞) 推薦系統zh_TW
dc.subject (關鍵詞) 行為經濟學交互推薦系統zh_TW
dc.subject (關鍵詞) 行為經濟學zh_TW
dc.subject (關鍵詞) 個人化zh_TW
dc.subject (關鍵詞) Recommendation systemen_US
dc.subject (關鍵詞) BeCrossRecen_US
dc.subject (關鍵詞) Behavior economicsen_US
dc.subject (關鍵詞) Personalizeden_US
dc.title (題名) BeCross：融合個人行為偏誤的推薦系統建模方法zh_TW
dc.title (題名) BeCross: A Personalized Behavior Bias Modeling Layer for Deep Recommender Systemsen_US
dc.type (資料類型) thesisen_US
dc.relation.reference (參考文獻) [1]. Adomavicius, G., Bockstedt, J. C., Curley, S. P., & Zhang, J. (2013). Do recommender systems manipulate consumer preferences? A study of anchoring effects. Information Systems Research, 24(4), 956-975. [2]. Anil, R., Gadanho, S., Huang, D., Jacob, N., Li, Z., Lin, D., ... & Yan, Q. (2022). On the factory floor: ML engineering for industrial-scale ads recommendation models. arXiv preprint arXiv:2209.05310. [3]. Chen, J., Dong, H., Wang, X., Feng, F., Wang, M., & He, X. (2023). Bias and debias in recommender system: A survey and future directions. ACM Transactions on Information Systems, 41(3), 1-39. [4]. Chen, Q., Zhao, H., Li, W., Huang, P., & Ou, W. (2019, August). Behavior sequence transformer for e-commerce recommendation in alibaba. In Proceedings of the 1st international workshop on deep learning practice for high-dimensional sparse data (pp. 1-4). [5]. Cheng, H. T., Koc, L., Harmsen, J., Shaked, T., Chandra, T., Aradhye, H., ... & Shah, H. (2016, September). Wide & deep learning for recommender systems. In Proceedings of the 1st workshop on deep learning for recommender systems (pp. 7-10). [6]. Chiny, M., Chihab, M., Bencharef, O., & Chihab, Y. (2022). Netflix recommendation system based on TF-IDF and cosine similarity algorithms. no. Bml, 15-20. [7]. Clarke, K. A. (2005). The phantom menace: Omitted variable bias in econometric research. Conflict management and peace science, 22(4), 341-352. [8]. Covington, P., Adams, J., & Sargin, E. (2016, September). Deep neural networks for youtube recommendations. In Proceedings of the 10th ACM conference on recommender systems (pp. 191-198). [9]. Kingma, D. P., & Ba, J. (2015). Adam: A method for stochastic optimization. In Proceedings of the 3rd International Conference on Learning Representations (ICLR). [10]. Ge, Y., Xu, S., Liu, S., Fu, Z., Sun, F., & Zhang, Y. (2020, July). Learning personalized risk preferences for recommendation. In Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval (pp. 409-418). [11]. Harper, F. M., & Konstan, J. A. (2015). The movielens datasets: History and context. Acm transactions on interactive intelligent systems (tiis), 5(4), 1-19. [12]. He, X., Liao, L., Zhang, H., Nie, L., Hu, X., & Chua, T. S. (2017, April). Neural collaborative filtering. In Proceedings of the 26th international conference on world wide web (pp. 173-182). [13]. Hornik, K. (1991). Approximation capabilities of multilayer feedforward networks. Neural networks, 4(2), 251-257. [14]. Houlsby, N., Giurgiu, A., Jastrzebski, S., Morrone, B., De Laroussilhe, Q., Gesmundo, A., ... & Gelly, S. (2019, May). Parameter-efficient transfer learning for NLP. In International conference on machine learning (pp. 2790-2799). PMLR. [15]. Hrnjic, E., & Tomczak, N. (2019). Machine learning and behavioral economics for personalized choice architecture. arXiv preprint arXiv:1907.02100. [16]. Jesse, M., & Jannach, D. (2021). Digital nudging with recommender systems: Survey and future directions. Computers in Human Behavior Reports, 3, 100052. [17]. Kahneman, D., & Tversky, A. (2013). Prospect theory: An analysis of decision under risk. In Handbook of the fundamentals of financial decision making: Part I (pp. 99-127). [18]. Koren, Y., Bell, R., & Volinsky, C. (2009). Matrix factorization techniques for recommender systems. Computer, 42(8), 30-37. [19]. Lex, E., Kowald, D., Seitlinger, P., Tran, T. N. T., Felfernig, A., & Schedl, M. (2021). Psychology-informed recommender systems. Foundations and trends® in information retrieval, 15(2), 134-242. [20]. Naumov, M., Mudigere, D., Shi, H. J. M., Huang, J., Sundaraman, N., Park, J., ... & Smelyanskiy, M. (2019). Deep learning recommendation model for personalization and recommendation systems. arXiv preprint arXiv:1906.00091. [21]. Raza, S., Rahman, M., Kamawal, S., Toroghi, A., Raval, A., Navah, F., & Kazemeini, A. (2024). A comprehensive review of recommender systems: Transitioning from theory to practice. arXiv preprint arXiv:2407.13699. [22]. Soto, C. J., & Jackson, J. J. (2013). Five-factor model of personality. Journal of Research in Personality, 42, 1285-1302. [23]. Sparck Jones, K. (1972). A statistical interpretation of term specificity and its application in retrieval. Journal of documentation, 28(1), 11-21. [24]. Steck, H., Baltrunas, L., Elahi, E., Liang, D., Raimond, Y., & Basilico, J. (2021). Deep learning for recommender systems: A Netflix case study. AI magazine, 42(3), 7-18. [25]. Sun, F., Liu, J., Wu, J., Pei, C., Lin, X., Ou, W., & Jiang, P. (2019, November). BERT4Rec: Sequential recommendation with bidirectional encoder representations from transformer. In Proceedings of the 28th ACM international conference on information and knowledge management (pp. 1441-1450). [26]. Thaler, R. H. (2016). Behavioral economics: Past, present, and future. American economic review, 106(7), 1577-1600. [27]. Thaler, R. H., & Sunstein, C. R. (2009). Nudge: Improving decisions about health, wealth, and happiness. Penguin. [28]. Tkalcic, M., Kosir, A., & Tasic, J. (2011). Affective recommender systems: the role of emotions in recommender systems. In The RecSys 2011 Workshops-Decisions@ RecSys 2011 and UCERSTI-2: Human Decision Making in Recommender Systems; User-Centric Evaluation of Recommender Systems and Their Interfaces-2 (Vol. 811, pp. 9-13). CEUR-WS. org. [29]. Tversky, A., & Kahneman, D. (1973). Availability: A heuristic for judging frequency and probability. Cognitive psychology, 5(2), 207-232. [30]. Tversky, A., & Kahneman, D. (1974). Judgment under Uncertainty: Heuristics and Biases: Biases in judgments reveal some heuristics of thinking under uncertainty. science, 185(4157), 1124-1131. [31]. Tversky, A., & Kahneman, D. (1981). The framing of decisions and the psychology of choice. science, 211(4481), 453-458. [32]. Uber Engineering. (2022, March 23). Graph learning powering the Uber Eats graph ecosystem. Uber Blog. https://www.uber.com/en-TW/blog/uber-eats-graph-learning/ [33]. Wang, R., Fu, B., Fu, G., & Wang, M. (2017). Deep & cross network for ad click predictions. In Proceedings of the ADKDD'17 (pp. 1-7). [34]. Wang, R., Shivanna, R., Cheng, D., Jain, S., Lin, D., Hong, L., & Chi, E. (2021, April). Dcn v2: Improved deep & cross network and practical lessons for web-scale learning to rank systems. In Proceedings of the web conference 2021 (pp. 1785-1797). [35]. Wang, T., Brovman, Y. M., & Madhvanath, S. (2021). Personalized embedding-based e-commerce recommendations at ebay. arXiv preprint arXiv:2102.06156. [36]. Wang, Z., She, Q., & Zhang, J. (2021). Masknet: Introducing feature-wise multiplication to CTR ranking models by instance-guided mask. arXiv preprint arXiv:2102.07619. [37]. Xia, X., Eksombatchai, P., Pancha, N., Badani, D. D., Wang, P. W., Gu, N., ... & Zhai, A. (2023, August). Transact: Transformer-based realtime user action model for recommendation at pinterest. In Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (pp. 5249-5259). [38]. Xiao, Y., Zhang, Y., & Li, X. (2024, June). modeling variational anchoring effect for recommender systems. In 2024 IEEE Conference on Artificial Intelligence (CAI) (pp. 926-931). IEEE. [39]. Zhao, Z., Hong, L., Wei, L., Chen, J., Nath, A., Andrews, S., ... & Chi, E. (2019, September). Recommending what video to watch next: a multitask ranking system. In Proceedings of the 13th ACM conference on recommender systems (pp. 43-51). [40]. Zhou, G., Zhu, X., Song, C., Fan, Y., Zhu, H., Ma, X., ... & Gai, K. (2018, July). Deep interest network for click-through rate prediction. In Proceedings of the 24th ACM SIGKDD international conference on knowledge discovery & data mining (pp. 1059-1068).zh_TW