2026-04-17T21:54:20.788Z http://ezaposleni.singidunum.ac.rs/rest/sciNaucniRezultati/oai

ezaposleni.singidunum.ac.rs/rest/sciNaucniRezultati/oai:2:6301 2019-07-03T05:21:58Z 2

Automated Language-Independent Authorship Verification (for Indo-European Languages) 2019 http://ezaposleni.singidunum.ac.rs/rest/sciNaucniRezultati/oai/record/2/6301 https://onlinelibrary.wiley.com/doi/abs/10.1002/asi.24163?af=R S. Adamović В. Мишковиц М. Шарац М. Милосављевић М. Веиновић In this paper, we have examined automated language independent authorship verification using text examples in several representative Indo-European languages, in cases when the examined texts belong to an open set of authors, i.e. author is unknown. We have showcased the set of developed language-dependent and language-independent features, the model of training examples, consisting of pairs of equal features for known and unknown texts, and the appropriate method of authorship verification. The authorship verification accuracy greater than 90% has been accomplished via the application of stylometric methods on four different languages (English, Greek, Spanish and Dutch, while the verification for Dutch is slightly lower). For the multilingual case, the highest authorship verification accuracy using basic machine learning methods, over 90%, has been achieved by the application of kNN and SVM-SMO methods, using the feature selection method SVM-RFE. The improvement in authorship verification accuracy in multilingual cases, over 94%, has been accomplished via ensemble learning methods, with the MultiboostAB method being a bit more accurate, but the RandomForest is generally more appropriate. article https://doi.org/10.1002/asi.24163 70 8 858 871 2330-1643 Journal of the Association for Information Science and Technology (JASIST)