The number of e-learning platforms has grown in recent years due to advances in cloud
computing, ease of access to technology, and the trend of people to constantly improve
their knowledge and skills. For this reason, a great deal of research has been carried out
to improve the use of these platforms. This work focuses on two research streams. First,
Emotion-aware systems are based on the use of different tools to capture students’ emotions
in order to adapt the lessons to their mood. Secondly, the main goal of Learning Analytics
is to harness educational data sets to infer, create, and predict new information that helps
to improve learning process.
Once the basis of this project have been defined, its main development has consisted of
the implementation of a system capable of detecting the mood of students in the course and
during the performance of different activities. This system has been integrated into one of
the most used e-learning platforms: Moodle.
With the purpose of displaying the data collected by this system, it has been necessary
to implement a set of visualizations in two dashboards, designed for teachers and students
respectively. In the same manner, this data has been analyzed with Machine Learning
techniques to infer relations, outliers, or trends.
Finally, to take advantage of the capabilities of the implemented emotion detector, a new
version of Ewetasker, a semantic task automation platform, has been developed. Through
it, students are able to adapt the environment to their emotions, improving their comfort
when performing tasks.
To summarize, the aim of this project has been to improve students mood through the
different developments carried out, and consequently, their academic performance.