Intelligent Analysis in Text Authorship Identification

Abstract

The problem of text authorship identifi cation is considered. A generalized review of the current state of the problem is provided. A solution based on the modifi cation of ensemble machine learning methods is proposed. The possibilities of applying sophisticated computational methods such as natural language processing, machine learning algorithms and stylometric analysis to identify individual writers based on their distinctive linguistic models are investigated. A hypothesis is put forward about the need for multidimensional text analysis and a new approach is proposed for identifying the authorship of the text in the form of a hybrid intelligent system.