Predicting school dropout rates is an important issue for the smooth execution of an educational system. This problem is solved by classifying students into two classes using educational activities related statistical datasets. One of the classes must identify the students who have the tendency to persist. The other class must identify the students who have the tendency to dropout. This problem often encounters a phenomenon that masks out the obtained results. This study delves into this phenomenon and provides a reliable educational data mining technique that accurately predicts the dropout rates. In particular, the three data classifying techniques, namely, decision tree, neural networks and Balanced Bagging, are used. The performances of these classifies are tested with and without the use of a downsample, SMOTE and ADASYN data balancing. It is found that among other parameters geometric mean and UAR provides reliable results while predicting the dropout rates using Balanced Bagging classifying techniques.

The school dropout problem has been recurrent in different educational areas, which has reinforced important challenges when pursuing education objectives. In this scenario, technical schools have also suffered from considerable dropout levels, even when considering a still increasing need for professionals in areas associated to computing and engineering. Actually, the dropout phenomenon may be not uniform and thus it has become urgent the identification of the profile of those students, putting in evidence techniques such as eXplainable Artificial Intelligence (XAI) that can ensure more ethical, transparent, and auditable use of educational data. Therefore, this article applies and evaluates XAI methods to predict students in school dropout situation, considering a database of students from the Federal Institute of Rio Grande do Norte (IFRN), a Brazilian technical school. For that, a checklist was created comprising explanatory evaluation metrics according to a broad literature review, resulting in the proposal of a new explainability index to evaluate XAI frameworks. Doing so, we expect to support the adoption of XAI models to better understand school-related data, supporting important research efforts in this area.

The main objective of this paper is to investigate how scientific modeling and visualization techniques can enhance the interpretation of educational data. For data modeling, k-means algorithm and correspondence analysis were used for clustering and dimensionality reduction, respectively. For data visualization, boxplot, violin chart and perceptual map techniques were analyzed. The analyses were performed using academic data from two courses of Instituto Federal do Rio Grande do Norte in the period from 2008 to 2016. One of the obtained results shows that students grades follow multimodal distributions, indicating that the use of classic boxplots is not adequate.