![]() | Only 14 pages are availabe for public view |
Abstract Authorship attribution analysis is a research field that assigns an author to anonymous text based on writing features. These features reflect the author’s gender, age, religion, education, job, motivation or ideology. It has several types of features such as character, lexical, syntactic, structural and semantic. Authorship attribution analysis solves different issues such discovering whether a specific text possesses to a specific author or not and detecting the most probable author for an unknown text which are called author verification. Moreover, extraction of gender, political view, religion, education, job or motivation from the author’s text is called author profiling. In addition, discrimination between two or more stylometric of authors according gender or political view or religion or education or job or motivation, revealing the right author for one disputed work between candidate authors. There are two issues are considered real social issues in the Islamic society. The first one is detecting the jurisprudence doctrine of anonymous fatwa. The second one is distinction between stylometric of jurisprudence doctrines in writing the fatwa.This thesis proposes using Arabic ontology as a semantic feature and n-gram as a lexical feature in the authorship attribution. These features are used for establishing a set of new proposed approaches. Moreover, it analyses the effect of changing the size of corpus using a fuzzy linguistic model (FLM) on the Arabic authorship attribution. The proposed approaches are split into two groups; each group solves a different issue in authorship attribution. The first group solves the problem of detecting the author of anonymous text (DAFLexical: Detecting the Author of the Fatwa using n-gram as a lexical feature, DAFLexFLM: Detecting the Author of the Fatwa using n-gram as a lexical feature and FLM, DAFSemantic: Detecting the Author of the Fatwa using Arabic ontology as a semantic feature, DAFSemFLM: Detecting the Author of the Fatwa using Arabic ontology as a semantic feature and FLM). |