Application of opinion mining techniques in the study of the perception of remote attendance in higher education institutions

Introduction: The COVID-19 pandemic had a significant impact on the educational sector, causing universities to adapt their processes to the so-called remote presence. In this sense, it became important for the universities to identify the perception of their members in order to provide feedback to the educational processes in the face of the potential arrival of a hybrid education model. Objective: Considering that affective computing is an area that attempts to obtain quantitative indicators from qualitative information, this article proposed the development of a study based on opinion mining on the responses to a perception survey with five qualitative questions, where a sample of students of Systems Engineering at the University of Cartagena were asked about the way in which the institution managed the academic processes during the pandemic. Methodology: Four methodological phases were defined for the development of the study: design and application of the instrument, cleaning and adequacy of the answers, analysis of polarity and generation of conclusions. Results: A polarity analysis was carried out on the responses provided, in which aspects referring to the development of academic activities during the pandemic were addressed. The study obtained as a result a ratio of 1.25 of positive polarity with respect to negative polarity over the total number of responses. Conclusions: The study made it possible to determine the relevance of opinion mining in the analysis of perception in educational contexts, as well as the possibility of extrapolating it to other areas such as marketing.