Abstracts Education

Web mining on e-learning

by Francisco José Casanova Faria Campos

Nowadays, the use of information technologies is vital in people’s lives and companies. This use and exchange of information generate a quantity of data that can create value for various sectors of society if possible to translate the information contained therein. Adopting this problem to the education sector, many issues and problems can be solved with the correct treatment of this data. The possibility of perceiving the students’ behaviours and study methodologies that allow greater school success is certainly a tempting opportunity and that can lead to the optimization of teaching as it is known and allow to create better people and professionals. Therefore, the focus of this project focuses on the processing of these data generated in university context, by creating of a solution that can receive different data types, apply analytical models, in order to generate reports and dashboards about the reality of the data In study, and predictive models, so that it can predict future grades of students based on their academic behaviour. In analytical terms it was possible to prove that there is a strong relationship between a great level of attendance and participation in classes with a good academic performance, that 77% of the students were able to take a final grade above 15 values, that all students were present at least one-third of the classes, in addition to proving the success of previously applied tools such as the card and rescue system, with 8 students, who had first failed at a critical evaluation moment, to be able to complete the course thanks to the latter mechanism. In predictive terms, the prototype proved to be effective, mainly in terms of regression, with an absolute error of 1.10 values. The data for this project were provided by the company IOTech, which gathered it from the use of students from different platforms in the “Web Programming” curriculum during the 2020/2021 school year. This work is also inserted in the IOScience project, carried out by the same company.