An improved framework for authorship identification in online messages

• Published in: Cluster computing Vol. 22; no. Suppl 5; pp. 12101 - 12110
• Main Authors: Srinivasan, L., Nalini, C.
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.09.2019; Springer Nature B.V
• Subjects: Algorithms; Authorship; Classification; Computer Communication Networks; Computer Science; Criminal law; Cybercrime; Electronic commerce; Identification; Internet; Machine learning; Messages; Operating Systems; Privacy; Processor Architectures; Social networks; Tracing; Virtual communities; Writing; Ngram features; Authorship identification; C4.5; Structural features; Lexical features; Adaboost and fuzzy classifier; Online messages; Syntactic features
• ISSN: 1386-7857; 1573-7543
• DOI: 10.1007/s10586-017-1563-3

The authorship identification will determine the likelihood of the writing produced, by an author, by means of examining the other writings. The rapid proliferation of technologies along with the applications of the internet, the misuse of online messages for the purpose of inappropriate or for illegal reasons is a major concern in society. The online message distribution and its anonymous nature will make the identity of tracing anyone of critical issue. The work has been developed using a framework for the identification of authorship of the online messages for addressing as well as tracing such problems. For this framework, identification of authorship is done by the four writing style features (the lexical, the syntactic, the structural, and the n-gram features) that are extracted and inductive learning algorithms have been used for building a feature based classification model for the identification of the authorship of the online messages. For this work, the C4.5, the fuzzy and also the Ada boost classifiers will be used for the task of authorship-identification. An experimental study on this framework with the effects of these classification techniques on online messages is evaluated.