ATRank: An Attention-Based User Behavior Modeling Framework for Recommendation 詳解
一、簡介 這是阿里巴巴和北大聯合發表的一篇論文，在電商環境中，用戶會存在多種行爲：瀏覽、購買、收藏商品，領取、使用優惠券，搜索關鍵詞等等。 由於attention具有抓取序列間的內在關係的能力，因此利用attention來對用戶不同的行爲進行建模，致力於更全面的融合用戶的不同行爲。 本文針對異構的行爲，提出的是一個通用框架，下游任務可以根據業務場景自行設計。 二、模型框架 模型主要分爲四個部分：
相關文章
- 1. Controllable Multi-Interest Framework for Recommendation
- 2. Practice on Long Sequential User Behavior Modeling for Click-Through Rate Prediction
- 3. 『 DSSM』A Multi-View Deep Learning Approach for Cross Domain User Modeling in Recommendation Systems
- 4. CoLink: An Unsupervised Framework for User Identity Linkage
- 5. Paper Notes: Adaptive User Modeling with Long and Short-Term Preferences for Personalized Recommenda
- 6. Graph Neural Networks for Social Recommendation
- 7. 『 論文閱讀』A Multi-View Deep Learning Approach for Cross Domain User Modeling in Recommendation Systems
- 8. IntentGC:a Scalable Graph Convolution Framework Fusing Heterogeneous Information for Recommendation
- 9. Generative Adversarial User Model for Reinforcement Learning Based Recommendation System
- 10. 文章淺析-《Joint Deep Modeling of Users and Items Using Reviews for Recommendation》
- 更多相關文章...
- • 免費ARP詳解 - TCP/IP教程
- • UDP報文格式詳解 - TCP/IP教程
- • Flink 數據傳輸及反壓詳解
- • 爲了進字節跳動，我精選了29道Java經典算法題，帶詳細講解
相關標籤/搜索
每日一句
-
每一个你不满意的现在，都有一个你没有努力的曾经。
歡迎關注本站公眾號,獲取更多信息
相關文章
- 1. Controllable Multi-Interest Framework for Recommendation
- 2. Practice on Long Sequential User Behavior Modeling for Click-Through Rate Prediction
- 3. 『 DSSM』A Multi-View Deep Learning Approach for Cross Domain User Modeling in Recommendation Systems
- 4. CoLink: An Unsupervised Framework for User Identity Linkage
- 5. Paper Notes: Adaptive User Modeling with Long and Short-Term Preferences for Personalized Recommenda
- 6. Graph Neural Networks for Social Recommendation
- 7. 『 論文閱讀』A Multi-View Deep Learning Approach for Cross Domain User Modeling in Recommendation Systems
- 8. IntentGC:a Scalable Graph Convolution Framework Fusing Heterogeneous Information for Recommendation
- 9. Generative Adversarial User Model for Reinforcement Learning Based Recommendation System
- 10. 文章淺析-《Joint Deep Modeling of Users and Items Using Reviews for Recommendation》