Thesis Topic Details

Topic ID:
3328
Title:
Incremental Discovery in Interaction Networks
Supervisor:
Mike Bain
Research Area:
Machine Learning, Bioinformatics
Associated Staff
Assessor:
Mark Temple
Topic Details
Status:
Active
Type:
R & D
Programs:
Group Suitable:
Industrial:
Pre-requisites:
--
Description:
Project overview:

With the growth of online applications in data-intensive areas like social
media, bioinformatics, etc. there is an increasing need to analyse and
mine interaction networks to discover interesting and potentially useful
patterns. This project will investigate rule-based and alternative
approaches to pattern discovery in interaction network graphs.

Approach:

Analysis of the structure of a number of interaction graphs from selected
data sets will be necessary. Some way will be needed to encode graph
'semantics' and basic features of the interaction structures in a
probabilistic rule-based form or related approach. A literature survey of
current algorithms and methods will be done at the start of the project.
Evaluation will be in terms of how well the system meets the design
requirements, as well as assessment of the outputs in domain terms.

Skills:

Programming in one of PHP, Perl, Javascript or Java, possibly database
design and implementation, simple user interface design and implementation.
Knowledge of rule-based approaches, both practise and theory, would be
useful.
Comments:
--
Past Student Reports
  Leyi ZHANG in s2, 2012
Incremental Discovery in Interaction Networks
 

Download report from the CSE Thesis Report Library

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