Educational institutions collect and manage information about their students in order to support decision-making aimed at improving academic performance and the overall achievement of their community. The objective of this research was to identify findings on academic performance considering sociocultural factors throughout secondary education. The study followed the KDD process (Knowledge Discovery in Databases) and applied machine learning algorithms for data mining. As a result, two main findings were identified: first, the family environment and the level of student responsibility directly influenced academic performance; second, five student profiles were detected based on these factors: (1) favorable family environment and high level of responsibility, associated with outstanding performance; (2) favorable environment and low responsibility, with limited performance; (3) unfavorabl...