LINE: Large-scale Information Network Embedding（线：大规模信息网络嵌入）

ABSTRACT

This paper studies the problem of embedding very largeinformation networks into low-dimensional vector spaces,which is useful in many tasks such as visualization, nodeclassification, and link prediction. Most existing graph em-bedding methods do not scale for real world information networks which usually contain millions of nodes. In thispaper, we propose a novel network embedding method called the “LINE,” which is suitable for arbitrary types of information networks: undirected, directed, and/or weighted. Themethod optimizes a carefully designed objective functionthat preserves both the local and global network structures.An edge-sampling algorithm is proposed that addresses thelimitation of the classical stochastic gradient descent andimproves both the effectiveness and the efficiency of the inference. Empirical experiments prove the effectiveness ofthe LINE on a variety of real-world information networks,including language networks, social networks, and citation networks. The algorithm is very efficient, which is able to learn the embedding of a network with millions of vertices and billions of edges in a few hours on a typical single machine. The source code of the LINE is available online.

摘要

研究了将非常大的信息网络嵌入到低维向量空间的问题，在可视化、非解密和链路预测等方面具有重要的应用价值。现有的图嵌入方法大多不能适应真实世界中包含数百万个节点的信息网络。本文提出了一种新的网络嵌入方法“线”，它适用于任意类型的信息网络:无向网络、有向网络和/或加权网络。该方法优化了精心设计的目标函数，同时保留了局部和全局网络结构。提出了一种边缘采样算法，解决了经典随机梯度下降算法的局限性，并证明了该算法的有效性和有效性。实证实验证明了这条线在各种真实世界的信息网络上的有效性，包括语言网络、社交网络和引文网络。该算法非常有效，能够在数小时内在一台典型的单机上学习嵌入具有数百万个顶点和数十亿条边的网络。源代码可以在网上找到。