Unsupervised Decomposition of a Document into Authorial Components

We propose a novel unsupervised method for separating out distinct authorial components of a document. In particular, we show that, given a book artificially “munged” from two thematically similar biblical books, we can separate out the two constituent books almost perfectly. This allows us to automatically recapitulate many conclusions reached by Bible scholars over centuries of research. One of the key elements of our method is exploitation of differences in synonym choice by different authors.

• Type: Journal
• Year: 2011
• Where: ACL
• Authors: Moshe Koppel, Navot Akiva, Idan Dershowitz, Nachum Dershowitz
• Keywords: ACL 2011, Bible Scholars, Biblical Books, Computational Linguistics, Constituent