Heterogeneous information networks (HINs), involving diverse types of nodes and links, are ubiquitous in the real world, ranging from biological and bibliographic networks to social networks. With heterogeneous types of nodes and links, HINs are able to model complex interactions and immensely rich semantics in real-world scenarios. Thus, HIN analysis has emerged as a promising direction for many data mining tasks. With the surge of network embedding, numerous recent research has shifted towards HIN embedding, which aims to project the nodes into a low-dimensional space whilst preserving the structural and semantic properties of HINs. The learned low- dimensional embedding has been a de facto solution towards fundamental problems in various applications, such as node classification, link prediction and recommendation.
In this tutorial, we will give a survey on recent developments of heterogeneous information network embedding and its applications in real applications. It will include the following aspects: (1) basic concepts in this field; (2) shallow HIN embedding models; (3) deep HIN embedding; (4) applications; (5) future work. This tutorial shall help researchers and practitioners to share new techniques in network embeddings and give some illustrative guidance for real applications.Tutorial Materials(TBD) Tutors
Chuan Shi, Beijing University of Posts and Telecommunications
E-mail:shichuan at bupt.edu.cn
Yanfang Ye, Case Western Reserve University
E-mail:yanfang.ye at case.edu
The topics of this tutorial cover main research directions of social network analysis and graph machine learning; and the target audiences are those who are interesting in heterogeneous network mining, representation learning and analysis from both academia and industry. The expected audiences have background on data mining. It is better for them to know social network analysis.
Social network is a mainstream direction of data mining. Moreover, heterogeneous network embedding is becoming a hot topic, and it is also widely used in many applications. The participants will know the recent developments of this direction, and learn to apply them to solve real problems.Outline of the tutorial
In this tutorial, we will give a survey on heterogeneous information network (HIN) embeddings and its applications in real problems. It includes the following aspects: (1) basic concepts in this field; (2) shallow HIN embedding models; (3) deep HIN embedding; (4) applications; (5) future work. This tutorial shall help researchers and practitioners to share new techniques for identifying and analyzing relationships in networks that integrate multiple types or sources of information.
The tutorial will cover the main and important work in this field. We have collected most of work in my webpage: http://www.shichuan.org/HIN_topic.html. We will select some important and representative work to present in the tutorial.