Gender Statistical Analysis Applied for Identifying Style Patterns in English Academic Writing

The present paper addresses the problem of writing style patterns in the context of English Academic Writing. Stylometric analysis is used in order to extract the main characteristics obtained from the evaluation of articles written in well-known scientific journals such as Elsevier and Springer. The objective of the paper is to establish a pattern description of articles written in the same domain depending on the gender of the authors. Relevant prior written work upon the current subject reveal different characteristics of writing style of authors from different cultural orientation and gender. The paper describes the main characteristics taken into account for the clustering model when it comes to title, abstract and chapters’ construction within the analyzed articles. A short description of the algorithms and tools for clustering and space reduction is presented for further selecting the best combination for the proposed model. An additional statistical layer is added to the current clustering algorithms and space reduction for obtaining statistical proven results of usage. An aggregated structure model is conducted as a result of characteristics selection and processing for future work usage in gender analysis of scientific articles writing. Conclusions and withdrawn along with the future directions extracted from the current work. A database structure is proposed formed out of statistical calculated percentage of papers depending on the author gender. The relevance of the work can be well used as a guide line in writing scientific articles as the main musts in scientific writing are presented.