In recent years, there has been evidence of a growing interest on the part of universities to know in advance the academic performance of their students and allow them to establish timely strategies to avoid desertion and failure. One of the biggest challenges to predicting student performance is presented in the course “Programming Fundamentals” of Computer Science, Software Engineering, and Information Systems Engineering careers in Peruvian universities for high student dropout rates. The objective of this research was to explore the efficiency of Long-Short Term Memory Networks (LSTM) in the field of Educational Data Mining (EDM) to predict the academic performance of students during the seventh, eighth, twelfth, and sixteenth weeks of the academic semester, which allowed us to identify students at risk of failing the course. This research compares several predictive models, such as Deep Neural Network (DNN), Decision Tree (DT), Random Forest (RF), Logistic Regression (LR), Support Vector Classifier (SVM), and K-Nearest Neighbor (KNN). A major challenge machine learning algorithms face is a class imbalance in a dataset, resulting in over-fitting to the available data and, consequently, low accuracy. We use Generative Adversarial Networks (GAN) and Synthetic Minority Over-sampling Technique (SMOTE) to balance the data needed in our proposal. From the experimental results based on accuracy, precision, recall, and F1-Score, the superiority of our model is verified concerning a better classification, with 98.3% accuracy in week 8 using LSTM-GAN, followed by DNN-GAN with 98.1% accuracy.

Currently higher education institutions are generating a lot of data as a result of their internal processes, but the limiting human capacities for the processing and analysis of large volumes of data cause these institutions to make hasty decisions and without real support, so it is necessary to use computational tools such as analytical systems that allow to speed up decision making based on data. The type of study is mixed and applied with a quasi-experimental design. An analytical system based on eight stages was built, which are: user characterization, business requirements, storage, technological infrastructure, feature engineering, computational processing, visualization and decision making. The analytical system was validated with data from the SQL Server management system of the academic services office and fifteen Excel files from the admission office oriented to academic data in a higher educational institution. The use of various algorithms such as support vector machine, random forest, k nearest neighbours, and neural networks, among others, was applied. The evaluation of the results generated by these algorithms allowed us to identify the support vector machine algorithm with 94.21% accuracy.

The COVID-19 pandemic changed the teaching modality radically, which led to virtual education in institutions, which had to quickly adapt as well as students. The research was carried out with a non-experimental, descriptive and correlational quantitative approach, applied to a sample of 1804 students from a public university, two statistical studies were carried out, one supported by the academic database and the second through the application of a validated instrument to identify which factors have the greatest influence on the academic performance of students in the virtual modality. In the statistical analysis, it was found that academic performance improved in the virtual semester with respect to the face-to-face one; demographic factors: gender, age, study cycle, and career groups had a significant influence on academic performance. Finally, no significant influence was found between the technological and methodological dimensions with the academic performance of the virtual semester.