A Computational Linguistic Perspective of Authorship and Plagiarism

This book presents an investigation of the potentialities of employing computational linguistics and statistics in resolving cases of disputed authorship and suspected plagiarism. Much has been written about the ways of addressing doubtful texts with no clear clues of real authorship. It is no use to reduplicate what has already been achieved. Therefore, some major questions are posed related to the viability of certain computational programs (WordSmith Tools, Version 5.0), and statistical techniques (Principal Components Analysis, and Cluster Analysis) involved in the statistical package (SPSS version 14.0). Corpora of various genres are tackled in English and Arabic so that some general premises might be attained. The statistical behavior of one particular feature is observed to pin down authorship and plagiarism problems. This behavior is related to the way function words behave across different textual corpora. The researcher figures out the overlapping aspects that bring plagiarism detection and authorship attribution together in one basket. Suspected cases of plagiarism can be interpreted as special cases of disputed or misattributed authorship.