A Model for Identifying Steps in Undergraduate Thesis Methodology

. Knowledge generation is an important asset of great economic powers, and knowledge societies are a fundamental part in the development of countries. Mexico is a country that is in the process of development and improvement of its education system, according to the Educational Reform promoted since 2012 by the Federal Government. We identified an area of opportunity at the undergraduate level to help improve the writing of students, specifically in draft theses and research proposals. This work focuses its efforts on analyzing with natural language processing techniques the ”Methodology” section, an important element for the development of a thesis, that helps the reader to understand if the techniques and data used are appropriate in an investigation. This paper proposes a Model to identify a series of steps in such a section. In addition, preliminary results of a basic exploration of a collected corpus are presented, pre-processing the text to generate a representation according to Language Models. The corpus contains documents of graduate and undergraduate levels in the computer science and information technologies domain. The preliminary results showed that the information extracted from the corpus serves to adequately differentiate the methodologies of both levels.