THE LUCS-KDD GROUP

(LIVERPOOL UNIVERSITY COMPUTER SCIENCE - KNOWLEDGE DISCOVERY IN DATA)

PhDs IN DATAMINING AND KDD

Liverpool University

Frans Coenen

Department of Computer Science

The University of Liverpool

Getting a PhD in Computer Science) - Doing Good Research, presented at BCS-SGAI (British Computer Society Specialist Group in AI) Student Event, Cambridge (Monday 10 December 2007).

1. SUGGESTED PhD PROGRAMMES OF RESEARCH

We are interested in finding applicants to undertake the following PhD programmes of research:

  1. 3-D Image Classification: A Study Using "Eye Ball" Data. The aim of the proposed research is to investigate mechanisms where by data mining techniques, specifically classification techniques, can be applied to 3-D images such as retinopathy volumes.
  2. Effective Candidate Sub-Graph Generation for Frequent Subgraph Mining. Is it possible to generate candidate frequent sub-graphs in such a way that duplication can be either avoided or minimised, and if so can the efficiency of current graph mining algorithms be enhanced?
  3. The Application of AI to Correct Tool Path Trajectories in Sheet Metal Forming. Can we AI techniques be applied to identify and apply corrections to tool path trajectories as used in the sheet metal forming industry? Projet would run in parallel with a current European Framework 7 project.
  4. Muti-agent Based Simulation for Multi-entity modelling: A Study in Predictive Biology. A study into the formal structures and processes required to support multi-entity based simulations for predictive modelling. Thetraget application will be the modeling of mouse populations so as to obtain a new understanding of their interactions
  5. Classification of Google Earth Entities.. An investigation into how data miming techniques can be applied to satellite image data in order to identify, group and classify features within satellite imagery.
  6. Volumetric Data Classification: A Study Directed at MRI Scan Imagery. Can we develop appropriate volumetric classification techniques, that incorporate 3-D spatial relationship information, while at the same time achieving effective performance given the significant size of volumetric data sets.

2. FURTHER POTENTIAL RESERACH TOPICS

In addition to the above the LUCS-KDD research group carry out research activities in many areas of KDD and Machine Learning. In particular the group is interested in attracting high calibre PhD students interested in carry out research work in the following areas:

Algorithms and techniques

  • Classification
  • Clustering
  • Frequent patterns
  • Rule discovery
  • Statistical techniques and mixture models
  • Constraint-based mining
  • Incremental algorithms
  • Scalable algorithms
  • Distributed and parallel algorithms
  • Privacy preserving data mining
  • Multi-relational data mining

Data mining and databases

  • Database integration
  • Inductive databases
  • Data mining query languages
  • Data mining query optimization

Data pre-processing

  • Dimensionality reduction
  • Data reduction
  • Discretization
  • Uncertain and missing information handling

Foundations of data mining

  • Complexity issues
  • Knowledge (pattern) representation
  • Global vs. local patterns
  • Logic for data mining
  • Statistical inference and probabilistic modelling

Innovative applications

  • Mining bio-medical data
  • Web content, structure and usage mining
  • Semantic web mining
  • Mining governmental data, mining for the public administration
  • Personalization
  • Adaptive data mining architectures
  • Invisible data mining

KDD process and process-centric data mining

  • Models of the KDD process
  • Standards for the KDD process
  • Background knowledge integration
  • Collaborative data mining
  • Vertical data mining environments

Mining different forms of data

  • Graph, tree, sequence mining
  • Semi-structured and XML data mining
  • Text mining
  • Temporal, spatial, and spatio-temporal data mining
  • Data stream mining
  • Multimedia miningPattern post-processing

Pattern post-processing

  • Quality assessment
  • Visualization
  • Knowledge interpretation and use

If you are interested in any of the above feel free to contact me at coenen@liverpool.ac.uk.

3. FUNDING AND THE APPLICATION PROCESS

Please note that, unfortunately, the LUCS-KDD group has insufficient funding currently to support PhD students. We might be able to find you some work within the department tutoring groups of students. This payes approximately £350 (UK pounds) per module (for roughly 30 hours work), most PhD students do 2 or 3 modules per year depending on avialability. In other words anything you earn within the department will not be sufficient to support your PhD studies! Thus you will have to provide your own funding --- sorry! For advise on scholarships etc. see:

http://www.liv.ac.uk/study/postgraduate/money/

A formal application will eventually be required. To do this you should go to:

http://www.csc.liv.ac.uk/research/pgresearch.html.

This www page also gives further information.

Within the application form be sure to include the phrase "supervisor = Frans Coenen" --- this will ensure that the application comes to me!



Created and maintained by Frans Coenen. Last updated 21 February 2011