CLRC: Computer Learning Research Centre return home buttonAbout Us buttonPeople buttonEvents buttonResearch buttonProjects button
Courses buttonPublications buttonResources buttonLinks buttonsearch buttonContact us button

Selected Research Grants: Alex Gammerman

Selected Grants:

2008-2011
Programme: EU FP7
Project Title: O-PTM-Biomarkers
Partners: King's College (London, UK); University of Copenhagen (Denmark); Transgene (France); UCL (UK); University of Porto (Portugal); University of Cologne (Germany); THYMED GmbH (Germany); ANTIGONE Ltd (UK)
Value: Total funding is £2.7 million
Objectives: Discovery of novel serum biomarkers based on aberrant post-translational modifications of O-glycoproteins, O-PTM-Biomarkers, and their application to early detection of cancer

2007-2010
Programme: VLA
Project Title: Development and Application of Machine Learning Algorithms for the Analysis of Complex Veterinary Data Sets
Value: £84,425

2007-2010
Programme:Cyprus Government
Project Title: Development of New Conformal Prediction Methods with Applications in Medical Diagnosis
Value: £30,770

2007-2010
Programme: EPSRC
Project Title: Practical Competive Prediction
Value: £406,854

2005-2008
Programme: MRC G0301107
Project Title: Proteomic Analysis of the Human Serum Proteome
Partners: UCL.
Value: £959,954
Objectives: To establish the potential of the human serum proteome for screening and diagnosis in large populations using existing cutting-edge technologies for high-throughput protein analysis and biomarker discovery.

Past Grants:

2003-2005
Programme: The Royal Society
Project Title: Efficient Pseudo-Random number generators.
Partners: Novosibirsk State University
Value: £7,800
Objectives: To find efficient procedures for random and and pseudo-random number generators.

2001-2004
Programme: EPSRC Project
Title: Complexity Approximation Principle and Predictive Complexity: Analysis and Applications
Partners: Royal Holloway, University of London
Value: £142,996
Objective: To developimplementation of the Complexity Approximation Principle and to compare thir performance with other learning algorithms; to study properties of predictive complexity in general and variants of predictive complexity for particular games of interest; theoretical analysis of conditional predictive complexity and predictive information; to investigate the connections between predictive complexity and randomness.

2001-2004
Programme: BBSRC
Project Title: Pattern recognition techniques for gene and promoter identification and classification in plant genomic sequences
Partners: Royal Holloway, University of London
Value: £150,648
Objective: To develop plant-specific gene prediction algorithms, based on the SVM technique; to compare the SVM approach with conventional techniques such as discriminant analysis or Hidden Markov Models in terms of accuracy and confidence of prediction and to develop SVM variants adapted to utilize sequence characteristics; to improve techniques for promoter and regulatroy region recognition and provide explicit candidates for these regions; to investigate classification of genes based on promoter sequences.

2000-2003
Programme:
BH Vector
Project Title: Support vector machines and related technologies
Partners: Royal Holloway, University of London
Value: $330,000
Objectives: To develop medical applications of Support Vector Machine and Transductive Confidence Machine

2000-2002
Programme:
E.U.
Project Title: The Development and Evaluation of New Methods for Editing and Imputation (EUREDIT)
Partners: Office of National Statistics (UK) University of York (UK) The Numerical Algorithms Group Limited (UK) Centraal Bureau voor de Statistiek (CBS) (Netherlands> Tilastokeskus (Statistics Finland) (Finland) Swiss Federal Statistical Office (SFSO) (Switzerland) Insiders Wissenbasierte Systeme GmbH (now replaced by (Germany) Istituto Nazionale Di Statistica (ISTAT) (Spain) Royal Holloway University of London (UK) University of Southampton (UK) University of Jyvaeskylae (JyU) (Finland) Statistics Denmark (DSt) (Denmark)
Value: £53,893
Objectives: To develop imputation methods with SVM

1999-2002
Programme: EPSRC - GR/M16856
Project Title: Comparison of the Support Vector Machine and Minimum Message Length Methods for Induction and Prediction
Partners: Royal Holloway, University of London
Value: £132,787
Objective: The theoretical analysis and comparison of performance of these two algorithms; the development of the MML method aimed at minimising a bound on prediction risk; and further algorithm development, testing and applications.

1997-2000
Programme: EPSRC - GR/L35812
Project Title: Support Vector and Bayesian Learning Algorithms
Partners: Royal Holloway, University of London
Value: £142,360
Objective: To develop algorithms for pattern recognition and regression.

1996-2000
Programme: IFR Ltd (formerly Marconi Instruments)
Project Title:Learning Algorithms for Fault Diagnosis
Partners: Royal Holloway, University of London
Value: £43,410
Objective: To develop a set of algorithms and software for fault diagnosis problems.

1996-1999
Programme: EU Framework IV
Project Title: Symbolic Data Analysis (SODAS)
Partners: Royal Holloway, University of London (UK), University of Paris (France), Office for National Statistics (UK), CDF Thompson (France) and another 14 partners from EU
Value:Total value of £4 millions; our part is £156,000
Objective: To develop a set of tools for symbolic data analysis.

RHUL navigation button Computer Science Department navigation button  
Last updated 18/12/07 9:32 / AGB