Authorship Identification System Using Word2Vec Word Embedding Model

The volume of text is growing rapidly, especially as a result of the publication of articles; the problem is made more difficult by the rise in text data that is anonymous. Authorship identification is a field of study where researchers investigate for various approaches to identify an unknown text's author. The aim is to determine who wrote anonymous texts by providing writing examples from potential authors. Authorship identification system identifies the most possible author of texts such as articles, news, books, blogs, emails and messages, etc. Authorship identification is the process of identifying the original author by scrutinizing a text's characteristics and writing style. Applications for this evaluation include news stories, forensic science, plagiarism detection, and responsibility for published work. Source code authorship identification has grown in significance over time due to the prevalence of online academic examinations, malware, and other types of code-based plagiarism. This system investigates how to identify the writers of English-language news articles. Extraction of features that indicate an author's writing style is the primary goal of the authorship identification challenge. In traditional methods of authorship identification, hand-crafted features are used to represent the text. Unlike traditional approaches, this system proposes the authorship identification by investigating the use of Word2Vec that performs automatic feature extraction. In the subject of authorship identification, deep layers of neural networks can employ word embedding to extract characteristics from them and learn the patterns of authors based on context and word co-occurrence.