Date of Publication

11-26-2008

Document Type

Master's Thesis

Degree Name

Master of Science in Computer Science

Subject Categories

Computer Sciences

College

College of Computer Studies

Department/Unit

Computer Science

Thesis Adviser

Rigan P. Ap-apid

Defense Panel Chair

Nelson Marcos

Defense Panel Member

Alexis V. Pantola
Rigan P. Ap-apid

Abstract/Summary

The advent of comprehensive metabolic profiles has facilitated the in-depth study of biochemical systems in their entirety. Aided by the effectiveness of the Biochemical Systems Theory (BST) modeling framework in numerically representing distinct features of such systems, the complex task of determining the behavior of each metabolite and its effect on the biochemical system is essentially reduced into a parameter estimation problem. An abundance of literatures have proposed the use of optimization algorithms to address the estimation task. Each approach asserts its own set of innovations and advantages over the rest; however, these claims are difficult to prove since qualitative data are generally biased while quantitative data are often incomparable and inconclusive due to dissimilar testing configurations. Hence, a standard benchmarking framework should be defined to serve as a neutral testing environment. Benchmark data collected from the optimization algorithm based on Interval Analysis and Constraint Propagation was extensively analyzed and may be compared with similar researches to provide a reliable point of comparison.

Abstract Format

html

Language

English

Format

Electronic

Accession Number

TG04568; CDTG004568

Shelf Location

Archives, The Learning Commons, 12F Henry Sy Sr. Hall

Physical Description

v, 58 leaves

Keywords

Parameter estimation; Interval analysis (Mathematics)

Upload Full Text

wf_yes

Download

Share

COinS