【SR-GNN】Session-based Recommendation with Graph Neural Networks詳解
結構圖： 首先，將所有會話序列建模爲定向會話圖，其中每個會話序列可以作爲子圖處理。然後依次處理每個會話圖，通過門控圖神經網絡得到每個圖中所有節點的潛在向量。然後，我們將每個會話表示爲全局首選項和用戶當前對該會話的興趣的組合，其中這些全局和本地會話嵌入向量都由節點的潛在向量組成。最後，對於每個會話，我們預測每個條目成爲下一個單擊的概率。在真實數據集上進行的大量實驗證明了該方法的有效性。 開源代碼在：
相關文章
- 1. GNN:Session-based Recommendation with Graph Neural Networks簡介
- 2. Graph Neural Networks for Social Recommendation
- 3. Memory Augmented Graph Neural Networks for Sequential Recommendation翻譯
- 4. Few-shot Learning with Graph Neural Networks
- 5. Few-Shot Learning with Graph Neural Networks
- 6. Overlapping Community Detection with Graph Neural Networks
- 7. Structural Image Classification with Graph Neural Networks
- 8. IMAGE DENOISING WITH GRAPH-CONVOLUTIONAL NEURAL NETWORKS
- 9. Graph-based Dependency Parsing with Graph Neural Networks
- 10. 筆記《Graph Neural Tangent Kernel: Fusing Graph Neural Networks with Graph Kernels》-NeurIPS 2019
- 更多相關文章...
- • 免費ARP詳解 - TCP/IP教程
- • *.hbm.xml映射文件詳解 - Hibernate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. GNN:Session-based Recommendation with Graph Neural Networks簡介
- 2. Graph Neural Networks for Social Recommendation
- 3. Memory Augmented Graph Neural Networks for Sequential Recommendation翻譯
- 4. Few-shot Learning with Graph Neural Networks
- 5. Few-Shot Learning with Graph Neural Networks
- 6. Overlapping Community Detection with Graph Neural Networks
- 7. Structural Image Classification with Graph Neural Networks
- 8. IMAGE DENOISING WITH GRAPH-CONVOLUTIONAL NEURAL NETWORKS
- 9. Graph-based Dependency Parsing with Graph Neural Networks
- 10. 筆記《Graph Neural Tangent Kernel: Fusing Graph Neural Networks with Graph Kernels》-NeurIPS 2019