Authorship Verification based on Linguistic Features

International Journal for Research in Applied Science and Engineering Technology, 2018

Every person is unique", we have been hearing this since ages. Every person has a unique identity, a unique fingerprint, a unique retina and a lot more. These features play a vital role in identification of individuals for security purposes. Unfortunately, when it comes to security of written pieces or words from an individual, these primary unique identities are futile. One cannot identify a writer from a written piece of text on the basis of retina or fingerprint scans, sometimes even the signature can be forged, in such situations for security purposes and intellectual property rights it becomes very important to identify the true author. Stylometry plays an important role in this. Every author has a unique style of writing, measure of this style of writing is called Stylometry. This paper proposes to identify authors from text based on their style of writing. First a data set consisting of articles, short stories and emails will be used to train the system for multiple authors, then a random text would be given to the system to identify the author correctly, if the author predicted by the system is similar to the author claimed then the information is authentic otherwise the author claiming to be the writer is a fraud. For stylometry, over the ages, many features have been focused on, but this paper proposes new features to be used for this purpose. While writing, there are many unconscious styles that are incorporated by the author, these features have been unnoticed till date, but can play a vital role in accurate and fast identification of authors. These features include: 'intellectual property right', 'chapter length', 'the importance of a word with respect to the other words in a document' and frequency of particular words per thousand words. The algorithms used to train the system can be Decision tree, Naive Bayesian or Multilayer Perceptron.

Scientific Programming

In this paper, a novel approach is presented for authorship identification in English and Urdu text using the LDA model with n-grams texts of authors and cosine similarity. The proposed approach uses similarity metrics to identify various learned representations of stylometric features and uses them to identify the writing style of a particular author. The proposed LDA-based approach emphasizes instance-based and profile-based classifications of an author’s text. Here, LDA suitably handles high-dimensional and sparse data by allowing more expressive representation of text. The presented approach is an unsupervised computational methodology that can handle the heterogeneity of the dataset, diversity in writing, and the inherent ambiguity of the Urdu language. A large corpus has been used for performance testing of the presented approach. The results of experiments show superiority of the proposed approach over the state-of-the-art representations and other algorithms used for authors...

SEEU Review

Authorship Analysis (AA) is a natural language processing field that examines the previous works of writers to identify the author of a text based on its features. Studies in authorship analysis include authorship identification, authorship profiling, and authorship verification. Due to its relevance, to many applications in this field attention has been paid. It is widely used in the attribution of historical literature. Other applications include legal linguistics, criminal law, forensic investigations, and computer forensics. This paper aims to provide an overview of the work done and the techniques applied in the authorship analysis domain. The examination of recent developments in this field is the principal focus. Many different criteria can be used to define a writer’s style. This paper investigates stylometric features in different author-related tasks, including lexical, syntactic, semantic, structural, and content-specific ones. A lot of classification methods have been ap...

2021

Constant technological growth has resulted in the danger and seriousness of cyber-attacks, which has recently unmistakably developed in various institutions that have complex Information Technology (IT) infrastructure. For instance, for the last three (3) years, the most horrendous instances of cybercrimes were perceived globally with enormous information breaks, fake news spreading, cyberbullying, crypto-jacking, and cloud computing services. To this end, various agencies improvised techniques to curb this vice and bring perpetrators, both real and perceived, to book in relation to such serious cybersecurity issues. Consequently, Forensic Writer Identification was introduced as one of the most effective remedies to the concerned issue through a stylometry application. Indeed, the Forensic Writer Identification is a complex forensic science technology that utilizes Artificial Intelligence (AI) technology to safeguard, recognize proof, extraction, and documentation of the computer or digital explicit proof that can be utilized by the official courtroom, especially, the investigative officers in case of a criminal issue or just for data analytics. This research's fundamental objective was to scrutinize Forensic Writer Identification technology aspects in twitter authorship analytics of various users globally and apply it to reduce the time to find criminals by providing the Police with the most accurate methodology. As well as compare the accuracy of different techniques. The report shall analytically follow a logical literature review that observes the vital text analysis techniques. Additionally, the research applied agile text mining methodology to extract and

2018

INTERNATIONAL JOURNAL OF CREATIVE RESEARCH THOUGHTS - IJCRT (IJCRT.ORG), 2022

As we know that each author has its own writing habits and certain inherent writing characteristics, that he or she uses in his or document writing unconsciously, we can study those hidden characteristics to identify their style of writing, this is very helpful in case of authorship detection, while two or more persons claim the ownership for the same content or document. Now a days, due to digitalization, someone can copy and paste your document and publish it with own name, which is a matter of copyright, but without sufficient proof, it is difficult to prove true legitimate ownership, in this paper we are discussing various characteristics and approaches and their importance that were used and discussed by many researchers in their work.

It is important to protect the identity of the works of an author. The content can be misused in many ways possible. Author identification is the solution for this problem. It analyses an author's writing style by extracting and studying features of his writing and comparing it with other documents. This paper proposes the implementation method for an author identification system. The proposed method consists of extraction of semantic features from the text, scoring these features and comparing them with a set of scores already stored previously in a database. This method is based upon the principles and concepts of text analysis.

International Journal of Engineering Applied Sciences and Technology, 2022

As digital media is growing, the problem of text proliferation is becoming a big problem. Therefore, the identification of true ownership of a document has become a cumbersome task. In the digital era, it is very easy to copy someone's document and publish it in their name. So it is very necessary to check the true authorship. Authorship attribution becomes difficult when we conduct it manually. However, this process needs automation, when the document size becomes large. AA is a mixture of art, science and technology that helps to discover the genuine authorship of an unknown text/document, based on its specific writing features. These specific features can reflect the author's mood, education, gender, age, ideology, religion, or motivation. Many kinds of characteristics, like lexical, character, structural, syntactic, and semantic are used in authorship recognition. In this experiment, we used approximately 120 different kinds of feature sets. In our experiment, we examined that the logistic classifier was working well and gave good results in the form of accuracy.

2016

In this paper, we deal with the author identification issues of the document whose origin is unknown. To overcome these problems, we propose a new hybrid approach combining the statistical and stylistic analysis. Our introduced method is based on determining the lexical and syntactic features of the written text in order to identify the author of the document. These features are explored to build a machine learning process. We obtained promising results by relying on PAN@CLEF2014 English literature corpus. The experimental results are comparable to those obtained by the best state of the art methods.

International Journal of Engineering and Advanced Technology, 2019

The internet is increasing exponentially with textual content primarily through social websites. The problems were also increasing with anonymous textual data in the internet. The researchers are searching for alternative techniques to know the author of an unknown document. Authorship Attribution is one such technique to predict the details of an unknown document. The researchers extracted various classes of stylistic features like character, lexical, syntactic, structural, content and semantic features to distinguish the authors writing style. In this work, the experiment performed with most frequent content specific features, n-grams of character, word and POS tags. A standard dataset is used for experimentation and identified that the combination of content based and n-gram features achieved best accuracy for prediction of author. Two standard classification algorithms were used for author prediction. The Random forest classifier attained best accuracy for prediction of author w...