A Stylometric Fingerprinting Method for Author Identification Using Machine Learning

Abstract

Identifying authors depends upon their writing patterns, known as stylometry. These writing patterns may differ for each person. It may be possible to have multiple documents for the same author. Due to the similarity in writing styles, it becomes difficult to judge the real author. These similarities may include the same font style and language, leading to difficulty identifying the exact author. Machine learning techniques assist in identifying human attitudes in written documents. Human behavior analysis can increase privacy and security by identifying malicious users and malware programs. The independent system of behavior analysis is seen as a threat to privacy, especially for those people who want to conceal their identity. Authorship attribution is how we use stylometry techniques to identify authors of given multiple authors. We will use the stylometry code process to extract the number of programmers from a given database. Analysis of different datasets assists in checking whether the coding style of a program remains consistent or not.