O'Reilly logo

Mining Heterogeneous Information Networks by Yizhou Sun, Jiawei Han

Stay ahead with the world's most comprehensive technology and business learning platform.

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, tutorials, and more.

Start Free Trial

No credit card required

CHAPTER 5

Meta-Path-Based Relationship Prediction

In Chapter 4, we introduced a meta-path-based similarity measure, PathSim, for heterogeneous information networks. The concept of meta-path serves not only as a basis for similarity search but also as a key for mining and learning general heterogeneous networks with an arbitrary network schema, because this notion provides a way to guide us to systematically build link-based features. In this chapter, we examine a new mining task, relationship prediction in heterogeneous information networks, by exploring meta-path-based features.

5.1 OVERVIEW

Link prediction, that is, predicting the emergence of links in a network based on certain current or historical network information, has been a popular ...

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, interactive tutorials, and more.

Start Free Trial

No credit card required