Authorship Identification in Bengali Literature: a Comparative Analysis

COLING 2012, Mumbai, December 2012. Authorship Identi ation in Bengali Literature: a Comparative Analysis Tanmoy Chakraborty Department of Computer S ien e & Engineering Indian Institute of Te hnology, Kharagpur India its_tanmoy se.iitkgp.ernet.in Abstra t Stylometry is the study of the unique linguisti styles and writing behaviors of individuals. It belongs to the ore task of text ategorization like authorship identi ation, plagiarism dete tion et . Though reasonable number of studies have been ondu ted in English language, no major work has been done so far in Bengali. In this work, We will present a demonstration of authorship identi ation of the do uments written in Bengali. We adopt a set of ne-grained stylisti features for the analysis of the text and use them to develop two di erent models: statisti al similarity model onsisting of three measures and their ombination, and ma hine learning model with De ision Tree, Neural Network and SVM. Experimental results show that SVM outperforms other state-of-the-art methods after 10-fold ross validations. We also validate the relative importan e of ea h stylisti feature to show that some of them remain onsistently signi ant in every model used in this experiment.