You are here

Chemometric Applications to a Complex Classification Problem: Forensic Fire Debris Analysis

Download pdf | Full Screen View

Date Issued:
2013
Abstract/Description:
Fire debris analysis currently relies on visual pattern recognition of the total ion chromatograms, extracted ion profiles, and target compound chromatograms to identify the presence of an ignitable liquid according to the ASTM International E1618-10 standard method. For large data sets, this methodology can be time consuming and is a subjective method, the accuracy of which is dependent upon the skill and experience of the analyst. This research aimed to develop an automated classification method for large data sets and investigated the use of the total ion spectrum (TIS). The TIS is calculated by taking an average mass spectrum across the entire chromatographic range and has been shown to contain sufficient information content for the identification of ignitable liquids. The TIS of ignitable liquids and substrates, defined as common building materials and household furnishings, were compiled into model data sets. Cross-validation (CV) and fire debris samples, obtained from laboratory-scale and large-scale burns, were used to test the models. An automated classification method was developed using computational software, written in-house, that considers a multi-step classification scheme to detect ignitable liquid residues in fire debris samples and assign these to the classes defined in ASTM E1618-10. Classifications were made using linear discriminant analysis, quadratic discriminant analysis (QDA), and soft independent modeling of class analogy (SIMCA). Overall, the highest correct classification rates were achieved using QDA for the first step of the scheme and SIMCA for the remaining steps. In the first step of the classification scheme, correct classification rates of 95.3% and 89.2% were obtained for the CV test set and fire debris samples, respectively. Correct classifications rates of 100% were achieved for both data sets in the majority of the remaining steps which used SIMCA for classification. In this research, the first statistically valid error rates for fire debris analysis have been developed through cross-validation of large data sets. The error rates reduce the subjectivity associated with the current methods and provide a level of confidence in sample classification that does not currently exist in forensic fire debris analysis.
Title: Chemometric Applications to a Complex Classification Problem: Forensic Fire Debris Analysis.
29 views
15 downloads
Name(s): Waddell, Erin, Author
Sigman, Michael, Committee Chair
Belfield, Kevin, Committee Member
Campiglia, Andres, Committee Member
Yestrebsky, Cherie, Committee Member
Ni, Liqiang, Committee Member
University of Central Florida, Degree Grantor
Type of Resource: text
Date Issued: 2013
Publisher: University of Central Florida
Language(s): English
Abstract/Description: Fire debris analysis currently relies on visual pattern recognition of the total ion chromatograms, extracted ion profiles, and target compound chromatograms to identify the presence of an ignitable liquid according to the ASTM International E1618-10 standard method. For large data sets, this methodology can be time consuming and is a subjective method, the accuracy of which is dependent upon the skill and experience of the analyst. This research aimed to develop an automated classification method for large data sets and investigated the use of the total ion spectrum (TIS). The TIS is calculated by taking an average mass spectrum across the entire chromatographic range and has been shown to contain sufficient information content for the identification of ignitable liquids. The TIS of ignitable liquids and substrates, defined as common building materials and household furnishings, were compiled into model data sets. Cross-validation (CV) and fire debris samples, obtained from laboratory-scale and large-scale burns, were used to test the models. An automated classification method was developed using computational software, written in-house, that considers a multi-step classification scheme to detect ignitable liquid residues in fire debris samples and assign these to the classes defined in ASTM E1618-10. Classifications were made using linear discriminant analysis, quadratic discriminant analysis (QDA), and soft independent modeling of class analogy (SIMCA). Overall, the highest correct classification rates were achieved using QDA for the first step of the scheme and SIMCA for the remaining steps. In the first step of the classification scheme, correct classification rates of 95.3% and 89.2% were obtained for the CV test set and fire debris samples, respectively. Correct classifications rates of 100% were achieved for both data sets in the majority of the remaining steps which used SIMCA for classification. In this research, the first statistically valid error rates for fire debris analysis have been developed through cross-validation of large data sets. The error rates reduce the subjectivity associated with the current methods and provide a level of confidence in sample classification that does not currently exist in forensic fire debris analysis.
Identifier: CFE0004954 (IID), ucf:49586 (fedora)
Note(s): 2013-08-01
Ph.D.
Sciences, Chemistry
Doctoral
This record was generated from author submitted information.
Subject(s): forensic science -- fire debris analysis -- gas chromatography mass spectrometry -- chemometrics -- multivariate statistics -- discriminant analysis -- principal components analysis (PCA) -- soft independent modeling of class analogy (SIMCA) -- error rates -- cross-validation
Persistent Link to This Record: http://purl.flvc.org/ucf/fd/CFE0004954
Restrictions on Access: campus 2014-08-15
Host Institution: UCF

In Collections