Mining heterogeneous social networks for egocentric information abstraction

Cheng Te Li, Shou De Lin

Research output: Chapter in Book/Report/Conference proceedingChapter


Social network is a powerful data structure that allows the depiction of relationship information between entities. However, real-world social networks are sometimes too complex for human to pursue further analysis. In this work, an unsupervised mechanism is proposed for egocentric information abstraction in heterogeneous social networks. To achieve this goal, we propose a vector space representation for heterogeneous social networks to identify combination of relations as features and compute statistical dependencies as feature values. These features, either linear or eyelie, intend to capture the semantic information in the surrounding environment of the ego. Then we design three abstraction measures to distill representative and important information to construct the abstracted graphs for visual presentation. The evaluations conducted on a real world movie datasct and an artificial crime dataset demonstrate that the abstractions can indeed retain significant information and facilitate more accurate and efficient human analysis.

Original languageEnglish
Title of host publicationFrom Sociology to Computing in Social Networks
Subtitle of host publicationTheory, Foundations and Applications
PublisherSpringer Vienna
Number of pages24
ISBN (Print)9783709102930
Publication statusPublished - 2010 Dec 1

All Science Journal Classification (ASJC) codes

  • Computer Science(all)


Dive into the research topics of 'Mining heterogeneous social networks for egocentric information abstraction'. Together they form a unique fingerprint.

Cite this