Lost in Translation: Authorship Attribution using Frame Semantics

We investigate authorship attribution using classifiers based on frame semantics. The purpose is to discover whether adding semantic information to lexical and syntactic methods for authorship attribution will improve them, specifically to address the difficult problem of authorship attribution of translated texts. Our results suggest (i) that frame-based classifiers are usable for author attribution of both translated and untranslated texts; (ii) that frame-based classifiers generally perform worse than the baseline classifiers for untranslated texts, but (iii) perform as well as, or superior to the baseline classifiers on translated texts; (iv) that---contrary to current belief---naive classifiers based on lexical markers may perform tolerably on translated texts if the combination of author and translator is present in the training set of a classifier.

[1]  I.N. Bozkurt,et al.  Authorship attribution , 2007, 2007 22nd international symposium on computer and information sciences.

[2]  Shlomo Argamon,et al.  Interpreting Burrows's Delta: Geometric and Probabilistic Foundations , 2007, Lit. Linguistic Comput..

[3]  D. Holmes.  A Stylometric Analysis of Mormon Scripture and Related Texts , 1992 .

[4]  James Allan,et al.  A comparison of statistical significance tests for information retrieval evaluation , 2007, CIKM '07.

[5]  Walter Daelemans,et al.  The effect of author set size and data size in authorship attribution , 2011, Lit. Linguistic Comput..

[6]  F. Mosteller,et al.  Inference and Disputed Authorship: The Federalist , 1966 .

[7]  Shlomo Argamon,et al.  Authorship attribution in the wild , 2010, Lang. Resour. Evaluation.

[8]  Craig H. Martell,et al.  Author Attribution Evaluation with Novel Topic Cross-validation , 2010, KDIR.

[9]  Richard Johansson,et al.  LTH: Semantic Structure Extraction using Nonprojective Dependency Trees , 2007, Fourth International Workshop on Semantic Evaluations (SemEval-2007).

[10]  F. Yates.  Contingency Tables Involving Small Numbers and the χ2 Test , 1934 .

[11]  Josef Ruppenhofer,et al.  FrameNet II: Extended theory and practice , 2006 .

[12]  Q. Mcnemar.  Note on the sampling error of the difference between correlated proportions or percentages , 1947, Psychometrika.

[13]  Adriana Kovashka,et al.  Authorship Attribution Using Probabilistic Context-Free Grammars , 2010, ACL.

[14]  Noah A. Smith,et al.  Probabilistic Frame-Semantic Parsing , 2010, NAACL.

[15]  Shlomo Argamon,et al.  Computational methods in authorship attribution , 2009 .

[16]  Michael Gamon,et al.  Linguistic correlates of style: authorship classification with deep linguistic analysis features , 2004, COLING.

[17]  S. Sathiya Keerthi,et al.  Which Is the Best Multiclass SVM Method? An Empirical Study , 2005, Multiple Classifier Systems.

[18]  C. E. Veni Madhavan,et al.  Stopword Graphs and Authorship Attribution in Text Corpora , 2009, 2009 IEEE International Conference on Semantic Computing.

[19]  Efstathios Stamatatos,et al.  A survey of modern authorship attribution methods , 2009, J. Assoc. Inf. Sci. Technol..

[20]  Danielle S. McNamara,et al.  Analyzing Writing Styles with Coh-Metrix , 2006, FLAIRS.

[21]  Joseph Rudman,et al.  The State of Authorship Attribution Studies: Some Problems and Solutions , 1997, Comput. Humanit..

[22]  G. Bruce Schaalje,et al.  Comparative Power of Three Author-Attribution Techniques for Differentiating Authors , 1997 .

[23]  Shlomo Argamon,et al.  Computational methods in authorship attribution , 2009, J. Assoc. Inf. Sci. Technol..

[24]  G. Leech,et al.  Word Frequencies in Written and Spoken English: based on the British National Corpus , 2001 .