Weighted Meta-Path Embedding Learning for Heterogenous Information Networks

EasyChair Preprint 3416

Abstract

A low-dimensional embedding can be easily applied in the downstream tasks for network mining and analysis. In the meantime, the popular models of random walk-based network embedding are viewed as the form of matrix factorization, whose computational cost is very expensive. Moreover, mapping different types of nodes into one metric space may result in incompatibility. To cope with the two challenges above, a weighted meta-path embedding framework (WMPE) is proposed in this paper. On one hand, a nearly-linear approximate embedding approach is leveraged to reduce the computational cost. On the other hand, the meta-path and its weight are learned to integrate the incompatible semantics in the form of weighted combination. Experiment results show that WMPE is effective and outperforms the state-of-the-art baselines on two real-world datasets.

Keyphrases: Heterogeneous Information Networks, Meta-path Learning, Network Embedding, Representation Learning

@booklet{EasyChair:3416,
  author    = {Zhang Yongjun and Yang Xiaoping and Wang Liang},
  title     = {Weighted Meta-Path Embedding Learning for Heterogenous Information Networks},
  howpublished = {EasyChair Preprint 3416},
  year      = {EasyChair, 2020}}