Thesis Topic Details

Topic ID:
3320
Title:
Graph-oriented Interaction Mining
Supervisor:
Mike Bain
Research Area:
Machine Learning, Social Networking Analysis, Finance
Associated Staff
Assessor:
Xiongcai Cai
Topic Details
Status:
Active
Type:
R & D
Programs:
Group Suitable:
Industrial:
Pre-requisites:
--
Description:
Data generated from an explicit or implicit interaction graph is becoming
increasingly common, in areas such as financial markets, bioinformatics
and social media. Increasingly, data mining methods are becoming required
to handle specialised aspects of pattern discovery in such data, such as
the transaction flow across markets or influence trends in social
networks.

This project will select one or more problems from this area and build on
previous work to investigate new methods of interaction mining. Useful
starting points include detection of patterns of authority in social
graphs using methods such as PageRank or Reverse PageRank algorithms, or
link prediction using ideas from recommender systems.
Comments:
--
Past Student Reports
  Richard Alan WEISS in s2, 2012
Graph-oriented Interaction Mining
 

Download report from the CSE Thesis Report Library

NOTE: only current CSE students can login to view and select reports to download.