Study of the process of identifying the authorship of texts written in natural language

The object of the research is the process of identifying the authorship of a text using computer technologies with the application of machine learning. The full process of solving the problem from text preparation to evaluation of the results was considered. Identification of the authorship of a text is a very complex and time-consuming task that requires maximum attention. This is because the identification process always requires taking into account a very large number of different factors and information related to each specific author. As a result, various problems and errors related to the human factor may arise in the identification process, which may ultimately lead to a deterioration in the results obtained. The subject of the work is the methods and means of analyzing the process of identifying the authorship of a text using existing computer technologies. As part of the work, the authors have developed a web application for identifying the authorship of a text. The software application was written using machine learning technologies, has a user-friendly interface and an advanced error tracking system, and can recognize both text written by one author and that written in collaboration. The effectiveness of different types of machine learning models and data fitting tools is analyzed. Computer technologies for identifying the authorship of a text are defined. The main advantages of using computer technology to identify text authorship are: – Speed: computer algorithms can analyze large amounts of text in an extremely short period of time. – Objectivity: computer algorithms use only proven algorithms to analyze text features and are not subject to emotional influence or preconceived opinions during the analysis process. The result of the work is a web application for identifying the authorship of a text developed on the basis of research on the process of identifying the authorship of a text using computer technology.