[cs-talks] Master Thesis Defense: Baichuan Zhou, 12/7 @ 3pm, Hariri Seminar Room

Harrington, Jacob Walter jwharrin at bu.edu
Fri Dec 1 15:19:07 EST 2017


Baichuan Zhou, Master Thesis Defense
Thursday, December 7th at 3pm in Hariri Seminar Room

Abstract:
Building data models using graphs has been ubiquitous in various research areas such as social network analysis, biological network analysis and image/video processing. Graph mining, as a result, is playing a key role in areas of data mining. However, most graph mining methods suffer from the curse of dimensionality. Graph embedding aims to map the graph into a lower-dimension space while preserving the relative proximity between nodes, which often serves as the first step to process the graph data. The quality of graph embedding has a substantial impact on the graph mining algorithms that follow. In this survey, we will formally define graph embedding problems, review representative graph embedding approaches and compare their performance in link prediction. We will then summarize our results and provide suggestions on the directions of future study.

Committee:
Professor George Kollios
Professor Lorenzo Orecchia

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://cs-mailman.bu.edu/pipermail/cs-talks/attachments/20171201/7e0c1dd7/attachment.html>


More information about the cs-talks mailing list