Loading...
|
Please use this identifier to cite or link to this item:
https://nccur.lib.nccu.edu.tw/handle/140.119/154212
|
| Title: | PetFoodRec：寵物食品推薦的異構圖學習模型——考慮間接消費者影響
PetFoodRec：A Heterogeneous Graph Learning Model for Pet Food Recommendation with Indirect Consumer Influence |
| Authors: | 鄭詠筑
Zheng, Yong-Zhu |
| Contributors: | 李蔡彥
Li, Tsai-Yen
鄭詠筑
Zheng, Yong-Zhu |
| Keywords: | 推薦系統
圖神經網路
個性化推薦
Recommendation System
Graph Neural Networks
Personalized Recommendation |
| Date: | 2024 |
| Issue Date: | 2024-11-01 11:22:49 (UTC+8) |
| Abstract: | 隨著社會文化的發展，寵物已成為許多家庭中不可或缺的一員，這種趨勢引發了對寵物食品需求的多樣化和個性化。然而，傳統的推薦系統通常僅關注用戶與商品之間的直接互動，忽略了寵物作為間接消費者對決策的影響，導致推薦結果難以滿足寵物的個性化需求。為了解決這一問題，本研究提出了一個創新的考慮間接消費者的寵物食品推薦系統，此系統將寵物視為購買決策中的關鍵影響因子，從而突破了傳統僅考慮購買者與商品互動的推薦模式。
透過從寵物用品網站蒐集所需的寵物食品數據集，我們將資料構建成含有用戶、寵物及寵物食品節點的異構圖(heterogeneous graph)，以表達三者之間的複雜交互關係。此異構圖被用作模型輸入，並引入圖神經網路（Graph Neural Networks, GNNs）和分層注意力機制以提取深層特徵，旨在獲取學習後的節點特徵，從而提供更精準的個性化寵物食品推薦。
本研究實驗結果證明，考慮寵物這個間接消費者的資訊，並使用異構圖神經網路捕捉潛在關係，可以顯著提升推薦系統的性能，使其在寵物食品推薦任務上優於現有的一般商品推薦系統，實現真正意義上的個性化寵物食品推薦。
As society and culture evolve, pets have become indispensable members of many households, driving a growing demand for diverse and personalized pet food options. However, traditional recommendation systems typically focus only on direct interactions between users and products, overlooking the influence of pets as indirect consumers in decision-making. This often results in recommendations that fail to address the personalized needs of pets. To address this issue, this study proposes an innovative pet food recommendation system that considers pets as key influencers in purchasing decisions, thus breaking away from traditional recommendation models that only consider interactions between buyers and products.
By collecting the required pet food dataset from pet supply websites, we constructed the data as a heterogeneous graph comprising nodes representing users, pets, and pet foods, capturing the complex interactions among them. This heterogeneous graph serves as the model’s input, where Graph Neural Networks (GNNs) and hierarchical attention mechanisms are utilized to extract deep features, aiming to learn enriched node representations. These learned features enable more accurate and personalized pet food recommendations.
Experimental results from this study demonstrate that considering pets as indirect consumers and leveraging heterogeneous graph neural networks to capture latent relationships significantly enhances the performance of recommendation systems. This approach outperforms conventional product recommendation systems in the task of pet food recommendation, achieving truly personalized pet food recommendations. |
| Reference: | [1] Zicker, S. C. (2008). Evaluating pet foods: how confident are you when you recommend a commercial pet food?. Topics in companion animal medicine, 23(3), 121-126.
[2] Rendle, S., Freudenthaler, C., Gantner, Z., & Schmidt-Thieme, L. (2012). BPR: Bayesian personalized ranking from implicit feedback. arXiv preprint arXiv:1205.2618.
[3] 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).
[4] Hamilton, W., Ying, Z., & Leskovec, J. (2017). Inductive representation learning on large graphs. Advances in neural information processing systems, 30.
[5] Veličković, P., Cucurull, G., Casanova, A., Romero, A., Lio, P., & Bengio, Y. (2017). Graph attention networks. arXiv preprint arXiv:1710.10903.
[6] Schlichtkrull, M., Kipf, T. N., Bloem, P., Van Den Berg, R., Titov, I., & Welling, M. (2018). Modeling relational data with graph convolutional networks. In The semantic web: 15th international conference, ESWC 2018, Heraklion, Crete, Greece, June 3–7, 2018, proceedings 15 (pp. 593-607). Springer International Publishing.
[7] Dong, Y., Chawla, N. V., & Swami, A. (2017, August). metapath2vec: Scalable representation learning for heterogeneous networks. In Proceedings of the 23rd ACM SIGKDD international conference on knowledge discovery and data mining (pp. 135-144).
[8] Wang, X., Ji, H., Shi, C., Wang, B., Ye, Y., Cui, P., & Yu, P. S. (2019, May). Heterogeneous graph attention network. In The world wide web conference (pp. 2022-2032).
[9] Hu, Z., Dong, Y., Wang, K., & Sun, Y. (2020, April). Heterogeneous graph transformer. In Proceedings of the web conference 2020 (pp. 2704-2710).
[10] Ying, R., He, R., Chen, K., Eksombatchai, P., Hamilton, W. L., & Leskovec, J. (2018, July). Graph convolutional neural networks for web-scale recommender systems. In Proceedings of the 24th ACM SIGKDD international conference on knowledge discovery & data mining (pp. 974-983).
[11] Wang, X., He, X., Wang, M., Feng, F., & Chua, T. S. (2019, July). Neural graph collaborative filtering. In Proceedings of the 42nd international ACM SIGIR conference on Research and development in Information Retrieval (pp. 165-174). |
| Description: | 碩士
國立政治大學
資訊科學系
111753137 |
| Source URI: | http://thesis.lib.nccu.edu.tw/record/#G0111753137 |
| Data Type: | thesis |
| Appears in Collections: | [資訊科學系] 學位論文
|
Files in This Item:
| File |
Description |
Size | Format | |
|---|
| 313701.pdf | | 5273Kb | Adobe PDF | 0 | View/Open |
|
All items in 政大典藏 are protected by copyright, with all rights reserved.
|
