A knowledge-based spectroscopic assignment system
Loading...
Files
Date
1993
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Studying the energy level structure of different molecules by means of spectroscopy is one of the fields in physics and chemistry. MOSAA[superscript 1] developed in this thesisis a knowledge-based system which can assist researchers in the assignment of the peaks of the molecules in question by using the spectral information provided, along with a knowledge base containing known energy levels of the given molecules and spectroscopic assignment rules provided by physicists.
Rules together with their associated components (i.e. parameters, functions and properties) compose the knowledge base for methanol spectroscopy assignment. Rules and their components are written in plain text using special grammars. A compiler created using lex and yacc is used to translate a rule file into an internal knowledge base data structure. A total of 313 rules for spectroscopic assignment are defined.
The MOSAA system inference engine combines backward chaining and forward chaining as well as two special mechanisms 'TRY' and 'Restart'. This inference engine is written in C++, and was designed specifically to handle the "generate and test" process required for spectroscopic assignment.
An overview of the MOSAA implementation is given. This includes a description of the top level structure, preprocessing and user interface. MOSAA was tested using two methanol species spectra (Spectrum CD[subscript 3][supercript 18]OH in the 900-1100 cm[superscript -1] region and Spectrum CD[subscript 3][superscript 16]OH in the 915-1030 cm[superscript -1] region). This testing showed that MOSAA successfully assigns 11 series in spectrum CH§ 8 0H testing and 11 series in spectrum CD[subscript 3][superscript 16]OH testing.