Award details

Methods for modelling complex spectral, chemical and biological systems

Reference	BBS/E/F/02107052
Principal Investigator / Supervisor	Professor E K Kemsley
Co-Investigators / Co-Supervisors
Institution	Quadram Institute Bioscience
Department	Quadram Institute Bioscience Department
Funding type	Research
Value (£)	27,408
Status	Completed
Type	Institute Project
Start date	01/04/1997
End date	31/03/1998
Duration	12 months

Abstract

An important family of mathematical modelling techniques has emerged in recent years: the 'natural computation' (NC) methods, so-called because these approaches are themselves modelled on systems or processes found in nature. Amongst the most well-known of these are artificial neural networks (ANNs), artificial life (cellular automata, fractals) and genetic algorithms (GAs). NC methods share the following characteristics: they are non-parametric, non-linear and highly versatile. In this project, we aim to exploit this power and versatility by examining three key areas of NC methodology, applied to problems of primary interest at and beyond IFR: - open-category or asymmetric classification applications these are typified by adulteration detection problems;- significance and confidence issues in NC methods including means of validating NC solutions and detecting over-fit solutions;- generic protocols for using NC methods to model large-scale, non linear systems typical applications include soft-field image reconstruction; modelling bioreactor cultures.

Summary

unavailable

Committee	Closed Committee - Agri-food (AF)
Research Topics	X – not assigned to a current Research Topic
Research Priority	X – Research Priority information not available
Research Initiative	X - not in an Initiative
Funding Scheme	X – not Funded via a specific Funding Scheme

I accept the terms and conditions of use (opens in new window)

export PDF file

back to list new search