Gamification in university education enhances student engagement and motivation, while artificial intelligence and learning analytics enable analysis of performance patterns and prediction of academic outcomes, facilitating early intervention strategies. This study aimed to evaluate the feasibility of predicting student final examination performance using data collected through Wooclap gamified sessions combined with machine learning and statistical modeling approaches in engineering education. The research was conducted across multiple engineering faculties at Universidad Politécnica de Madrid (UPM), analyzing data from eight undergraduate and graduate courses. Ten structured Wooclap sessions were implemented in each course throughout the academic semester. The predictive modeling framework incorporated Random Forest machine learning algorithms and statistical models, considering variables including Wooclap session performance, attendance rates, and course-specific characteristics. Data analysis revealed significant correlations between early learning analytics indicators and final examination outcomes. The machine learning models showed promising capabilities in identifying students at risk of underperformance, with attendance patterns and interactive session results proving particularly informative for prediction accuracy. The results indicate substantial potential for implementing early warning systems in higher education, enabling timely academic interventions. The combination of gamification with predictive analytics provides dual benefits of enhancing student engagement while offering valuable insights for educational planning. These findings demonstrate practical applications of artificial intelligence in university settings and establish a scalable framework for similar predictive systems across diverse engineering disciplines.

Keywords: Learning Assessment, Gamification, Artificial Intelligence, Learning Analytics.