召回方法的调研

关于

本文收集了在搜索、广告、推荐系统中的一些召回策略和算法

协同过滤

基于内容的召回

基于用户群

倒排链

{ 
    'tag_1': 
        [ { itemID: '13', weight: 0.7 }, { itemID: '2', weight: 0.53 } ],
    'tag_2': 
        [ { itemID: '1', weight: 0.37 } ],
    ...
}

向量召回

参考资料

  1. https://cloud.tencent.com/developer/article/1174893
  2. J. Weston, A. Makadia, and H. Yee. Label partitioning for sublinear ranking. In S. Dasgupta and D. Mcallester, editors, Proceedings of the 30th International Conference on Machine Learning (ICML-13), volume 28, pages 181–189. JMLR
  3. T. Liu, A. W. Moore, A. Gray, and K. Yang. An investigation of practical approximate nearest neighbor algorithms. pages 825–832. MIT Press, 2004