It can be considered per form of style-based document authentication (Echtheitskritik), which has valuable applications that extend well beyond the domain of literary analysis, to, for instance, the domain of forensic sciences. According esatto Stamatatos’s 2009 survey of the field, ‘[t]he main ispirazione behind statistically or computationally-supported authorship attribution is that by measuring some textual features we can distinguish between texts written by different authors.’22 22 Ed. Stamatatos, ‘Per survey‘ (n. 14, above) 538. This basic assumption implies that it should be possible onesto assess, for any new unseen document, whether or not it was written by other authors for whom we have texts available. Nowadays computational authorship studies are often considered a subfield of stylometry per the digital humanities, the broader computational study of the writing style of texts.23 23 D. Holmes, ‘The evolution of stylometry sopra humanities scholarship‘, LLC 13 (1998) 111–17.
While stylometry has per rich history, dating back preciso at least the nineteenth century, it is clear that it received its most important impetus only in the past two or three decades, stimulated by the rise of (personal) computing and the increased availability of large bodies of text con electronic form. Apart from the influential, yet more conventional, statistical analyses carried out by pioneers such as Mosteller and Wallace or John Burrows well before the 1990s, an influential approach in authorship studies has been esatto approach the attribution of anonymous texts as verso ‘text categorization‘ problem.24 24 Mosteller and Wallace, Inference and disputed authorship (n. 4, above) and J. Burrows, Computation into criticism: per study of Jane Austen’s novels (Oxford 1987). Heavily influenced by parallel research durante calcolatore elettronico science, the timore was onesto optimize verso statistical classifier on example texts by verso number of available candidate authors, much like a spam filter nowadays is still trained on manually annotated emails esatto learn how onesto distinguish between ‘junk‘ email and normal messages.25 25 F. Sebastiani, ‘Machine learning per automated text categorisation‘, ACM Pc Surveys 34 (2002) 1–47. After preparazione such per classifier on this example scadenza, the classifier could then be used esatto categorize or classify anonymous text as belonging puro one of the istruzione authors‘ oeuvres.
It resembles per police lineup, con which the correct author of an anonymous text has to be singled out from a series of available candidate authors for whom reference or ‘training‘ material is available
This text categorization setup is commonly whiplr known as ‘authorship attribution‘.26 26 The following paragraph heavily draws on M. Koppel and Y. Winter, ‘Determining if two documents are written by the same author‘, JASIST 65 (2014) 178–187. For a number of years, practitioners of stylometry have che tipo di onesto acknowledge the limitations of authorship attribution, because it necessarily assumes that the correct target author is indeed included durante the arnesi of candidates. Sopra many real-world cases, this problematic assumption cannot possibly be made, because the arnesi of relevant candidates is difficult or impossible sicuro establish beforehand. Because of this, the setup of authorship verification has recently been introduced as per new framework: here, the task is esatto verify whether or not an anonymous document was written by one or several of verso series of candidate authors. In some sense, authorship verification redefines the text categorization problem by adding an additional category label: ‘None of the above.‘
Con the present context, it should be emphasized that the problem posed by the HA is per ‘vanilla‘ example of a problem durante authorship verification: while the corpo indeed contains per number of (auto-) attributions, the veracity of all of these has been questioned con previous scholarship
Verification is hence an increasingly common experimental setup per authorship studies, and is the topic of verso dedicated track mediante the yearly PAN competition, an annual competition on finding computational solutions preciso issues in present-day textual forensics, mostly related preciso the detection of plagiarism, authorship, and aimable programma misuse (such as grooming or Wikipedia vandalism).27 27 The competition’s website is pan.webis.de. The most recent survey of an authorship verification track is: Addirittura. Stamatatos et al., ‘Overview of the author identification task at PAN 2015′ con Working Taccuino Papers of the CLEF 2015 Evaluation Labs, di nuovo. L. Cappellato et al. (2015). Generally speaking, authorship verification is per more generic problem than authorship attribution – i.e. every attribution problem could, in principle, be cast as verso verification problem – but it has also proven puro be more challenging. Mediante our experiments, we have therefore attempted sicuro radically minimize any assumptions on our part as esatto the authorial provenance of the texts per the HA. For each piece of text analysed below, we propose onesto independently assess the probability that it was written by one of the (alleged) individual authors identified per the corpus.