论文阅读——LANE-Label Informed Attrib

2018-09-29  本文已影响0人  泽泽馥泽泽

带标签数据的sku嵌入方法

方法名:Label Informed Attributed Network Embedding

简称:LANE

sku嵌入向量中应包括:user对sku的行为,sku属性,sku标签

算法基本流程

LANE 细节

网络的构造

construct.png

算法伪代码

Algorithm :Label Informed Attributed Network Embedding
Input: d(嵌入维度)
Input: max\_iter(迭代次数)
Input: G(带权邻接矩阵)
Input: A(属性矩阵)
Input: \alpha _{1},\alpha_{2}(权重参数)

Output:H(sku嵌入矩阵)

设sku数量(即构造图中的节点数量)为n,sku属性的维度为m, sku标签的维度为k,sku嵌入向量维度为d

G \in R^{n*n}, A \in R^{n*m}, Y \in R^{n*k}

S^{(G)},S^{(A)} \in R^{(n*n)}

L^{(G)}, L^{(A)}, L^{(Y)} \in R^{n*n}

U^{(G)}, U^{(A)}, U^{(Y)},H \in R^{n*d}

1 : Construct the affinity matrices S^{(G)} and S^{(A)}
2 : Compute Laplacian matrices L^{(G)} , L^{(A)} and L^{(Y)}
3 : Initialize t = 1, U^{(A)}=0, U^{(Y)} =0,H=0
4 : repeat
5 :      Update U^{(G)}
(L^{(G)} + \alpha_{1} U^{(A)} U^{(A)^{T}} + \alpha_{2} U^{(Y)} U^{(Y)^{T}} + HH^{T})U^{(G)} = \lambda_{1}U^{(G)}
6 :      Update U^{(A)}
(\alpha_{1}L^{(A)} + \alpha_{1} U^{(G)} U^{(G)^{T}} + HH^{T})U^{(A)} = \lambda_{2}U^{(A)}
7 :      Update U^{(A)}
(\alpha_{2}L^{(YY)} + \alpha_{2} U^{(G)} U^{(G)^{T}} + HH^{T})U^{(Y)} = \lambda_{3}U^{(Y)}
8 :      Update H
(U^{(G)} U^{(G)^{T}} + U^{(A)} U^{(A)^{T}} + U^{(Y)} U^{(Y)^{T}})H = \lambda_{4}H
9 : t = t +1
10 : until max_iter
11 : return H

spark关键代码

上一篇下一篇

猜你喜欢

热点阅读