Algerian Journal of Forensic Sciences and Criminology
Volume 2, Numéro 2, Pages 12-16
2025-12-29
Comparative Analysis Of Ftir Results Of Textile Fibers Using Visual And Chemometric Methods
Authors : Naamane Abderrahmani . Tarakat Ali . Mohamed Cherif Cherfa .
Abstract
Forensic experts have traditionally relied on univariate statistical methods, such as comparing the refractive index of glass from known and unknown sources using Student's t-test or the Welch test. In contrast, multivariate statistical analysis, or chemometrics, has received less attention. Chemometrics can be applied to complex chemical data, including various types of spectra or chromatographic profiles. This approach utilizes multiple variables simultaneously to describe the structure of large datasets, enabling the detection of similarities and interpretation of differences between spectra. Historically, such comparisons have been performed visually by experts. However, a significant challenge arises when dealing with high-dimensional data, such as absorption spectra, which typically contain hundreds or even thousands of wavelengths. With large sample sizes, even experienced analysts can only reliably identify major peaks, while subtler variations within complex datasets are often impossible to discern without advanced statistical tools
Keywords
Statistical Analysis ; Multivariate Analysis ; Chemometrics ; Principal Component Analysis (PCA) ; Forensic science
Les articles similaires
Abid A.
.
Zouhri A.
.
Sennoune M.
.
pages 165-171.
Abdellaoui Khadidja
.
pages 42-53.
Sekkouti Yassine Hacene
.
Zinet Fatima
.
pages 939-952.
Amar Irekti1*, Mehena Oualit1, Fatima Zibouche1, 2 Amar
.
Oualit Mehena
.
Zibouche Fatima
.
Buncianu Dorel
.
pages 14-19.
Mohamed Lamine Amor
.
Abderrazak Lemrini
.
pages 90-97.
