Questioned Document Examination

Abstract

Questioned documents is a significant field of forensic science that deals with handwriting, printing, and typewriting analysis. Counterfeit documents are increasingly being produced using various means, necessitating expertise in questioned documents and analytical research methods. Non-destructive or quasi-destructive analysis methods are crucial, considering the nature of the examined documents, and these are discussed in this chapter. However, interpreting the spectral information obtained from modern spectrophotometers, which includes data on inks, toners, papers, and other materials in the questioned document, presents a significant challenge due to the sheer volume of data and chemical information involved. This poses a constant need for experts to analyse large datasets and extract meaningful information that can answer court queries and aid in accurate interpretation of outcomes. Chemometric methods, such as supervised and non-supervised approaches, are employed to reduce large datasets, systematically interpret results, and address classification and discrimination problems in questioned document examination. This chapter discusses several chemometric approaches, including principal component analysis (PCA), hierarchical clustering analysis (HCA), linear discriminant analysis (LDA), support vector machine (SVM), and others. The chapter presents multivariate statistical analysis as an effective method for extracting important spectral features necessary for discrimination and classification of documents, with a specific focus on inks, toners, and paper analysis. Overall, the chapter highlights the complexities and challenges involved in analysing questioned documents and the importance of utilising chemometric methods for data analysis and interpretation in this field of forensic science.