DSpace Repository

Deep Combination of Stylometry Features in Forensic Authorship Analysis

Show simple item record

dc.contributor.author Canbay, Pelin
dc.contributor.author Sezer, Ebru
dc.contributor.author Sever, Hayri
dc.date.accessioned 2021-06-17T11:50:57Z
dc.date.available 2021-06-17T11:50:57Z
dc.date.issued 2020
dc.identifier.citation Canbay, Pelin; Sezer, Ebru; Sever, Hayri (2020). "Deep Combination of Stylometry Features in Forensic Authorship Analysis", International Journal of Information Security Science, Vol. 9, no. 3, pp. 154-163. tr_TR
dc.identifier.issn 2147-0030
dc.identifier.uri http://hdl.handle.net/20.500.12416/4826
dc.description.abstract Authorship Analysis (AA) in forensic is a process aim to extract information about an author from his/her writings. Forensic AA is needed for detection characteristics of anonymous authors to make better the security of digital media users who are exposed to disturbing entries such as threats or harassment emails. To analyze whether two anonymous short texts were written by the same author, we propose a combination of stylometry features from different categories in different progress. In the majority of the previous AA studies, the stylometric features from different categories are concatenated in a preprocess. In these studies, during the learning process, no category-specific operations are performed; all categories used are evaluated equally. On the other hand, the proposed approach has a separate learning process for each feature category due to their qualitative and quantitative characteristics and combines these processes at the decision phase by using a Combination of Deep Neural Networks (C-DNN). To evaluate the Authorship Verification (AV) performance of the proposed approach, we designed and implemented a problem-specific Deep Neural Network (DNN) for each stylometry category we used. Experiments were conducted on two English public datasets. The results show that the proposed approach significantly improves the generalization ability and robustness of the solutions, and also have better accuracy than the single DNNs. tr_TR
dc.language.iso eng tr_TR
dc.rights info:eu-repo/semantics/openAccess tr_TR
dc.subject Forensic Authorship Analysis tr_TR
dc.subject Deep Neural Networks tr_TR
dc.subject Neural Network Combination tr_TR
dc.subject Anonymous Document Pairs tr_TR
dc.title Deep Combination of Stylometry Features in Forensic Authorship Analysis tr_TR
dc.type article tr_TR
dc.relation.journal International Journal of Information Security Science tr_TR
dc.contributor.authorID 11916 tr_TR
dc.identifier.volume 9 tr_TR
dc.identifier.issue 3 tr_TR
dc.identifier.startpage 154 tr_TR
dc.identifier.endpage 163 tr_TR
dc.contributor.department Çankaya Üniversitesi, Mühendislik Fakültesi, Yazılım Mühendisliği Bölümü tr_TR


Files in this item

This item appears in the following Collection(s)

Show simple item record