Abstract:

The statistical analysis of literary style is one of many fields which has benefited significantly from advances in computing technology. These days it is very easy to process large documents and extract numerous pieces of information from them. With so much data available, we need tools to help us visualize and make sense of the data. After outlining some landmark studies done in this field I will show an application of canonical discriminant analysis (CDA) and principal component analysis (PCA) to the problem of authorship testing and illustrate some the strengths and weaknesses of these techniques.