The GI therefore proposes the following iterative procedure, which can be likened preciso forms of ‘bootstrapping’

Let quantitativo represent an unknown document and let y represent verso random target author’s stylistic ‘profile’. During one hundred iterations, it will randomly select (a) fifty verso cent of the available stylistic features available (di nuovo.g. word frequencies) and (b) thirty distractor authors, or ‘impostors’ from a pool of similar texts. Con each iteration, the GI will compute whether interrogativo is closer to y than sicuro any of the profiles by the thirty impostors, given the random selection of stylistic features con that iteration. Instead of basing the verification of the direct (first-order) distance between interrogativo and y, the GI proposes esatto record the proportion of iterations per which quantitativo was indeed closer esatto y than onesto one of the distractors sampled. This proportion can be considered per second-order metric and will automatically be verso probability between nulla and one, indicating the robustness of the identification of the authors of quantitativo and y. Our previous rete di emittenti has already demonstrated that the GI system produces excellent verification results for classical Latin prose.31 31 Compagno the setup durante Stover, et al, ‘Computational authorship verification method’ (n. 27, above). Our verification code is publicly available from the following repository: This code is described durante: M. Kestemont et al. ‘Authenticating the writings’ (n. 29, above).

For modern documents, Koppel and Winter were even able onesto report encouraging scores for document sizes as small as 500 words

We have applied verso generic implementation of the GI puro the HA as follows: we split the individual lives into consecutive samples of 1000 words (i.di nuovo. space-free strings of alphabetic characters), after removing all punctuation.32 32 Previous research (see the publications mentioned mediante the previous two taccuino) suggests that 1,000 words is per reasonable document size con this context. Each of these samples was analysed individually by pairing it with the profile of one of the HA’s six alleged authors, including the profile consisting of the rest of the samples from its own text. We represented the sample (the ‘anonymous’ document) by verso vector comprising the imparfaite frequencies of the 10,000 most frequent tokens in the entire HA. For each author’s profile, we did the same, although the profile’s vector comprises the average incomplete frequency of the 10,000 words. Thus, the profiles would be the so-called ‘mean centroid’ of all individual document vectors for per particular author (excluding, of course, the current anonymous document).33 33 Koppel and Seidman, ‘Automatically identifying’ (n. 30, above). Note that the use of a single centroid a author aims puro veterano, at least partially, the skewed nature of our momento, since some authors are much more strongly represented sopra the campione or retroterra pool than others. If we were not using centroids but mere text segments, they would have been automaticallysampled more frequently than others during the imposter bootstrapping.

Sicuro the left, verso clustering has been added on culmine of the rows, reflecting which groups of samples behave similarly

Next, we ran the verification approach. During one hundred iterations, we would randomly select 5,000 of the available word frequencies. We would also randomly sample thirty impostors from a large ‘impostor pool’ of documents by Latin authors, including historical writers such as Suetonius and Livy.34 34 See Appendix 2 for the authors sampled. The pool of impostor texts can be inspected durante the code repository for this paper. Per each iteration, we would loveaholics check whether the anonymous document was closer sicuro the current author’s profile than onesto any of the impostors sampled. In this study, we use the ‘minmax’ metric, which was recently introduced con the context of the GI framework.35 35 See Koppel and Winter, ‘Determining if two documents’ (n. 26, above). For each combination of an anonymous text and one of the six target authors’ profiles, we would record the proportion of iterations (i.e. per probability between nulla and one) in which the anonymous document would indeed be attributed preciso the target author. The resulting probability table is given in full mediante the appendix preciso this paper. Although we present per more detailed tete-a-tete of this giorno below, we have added Figure 1 below as an intuitive visualization of the overall results of this approach. This is verso heatmap visualisation of the result of the GI algorithm for 1,000 word samples from the lives sopra the HA. Cell values (darker colours mean higher values) represent the probability of each sample being attributed preciso one of the alleged HA authors, rather than an imposter from per random selection of distractors.