Measuring Structural Distances between Texts.
Uli Fahrenberg, Fabrizio Biondi, Kevin Corre, Cyrille Jégourel, Simon Kongshøj, Axel Legay · arXiv (Cornell University) · 2014
We define and use a new inter-textual distance which, contrary to other common approaches, not only measures differences in occurrences of words, but in occurrences of multi-word phrases. We show that this distance may easily be calculated and use it for statistical analysis of some sample corpuses of genuine and fake scientific papers.