The education has a crucial role in the society; therefore, it is important to have quality education systems. However, the COVID-19 pandemic influenced education. Therefore, it is interesting to know to what extent education was affected due to this health crisis. In this sense, the academic achievement of students is an indicator that allows knowing the quality of educational systems. One way to measure this indicator is through the analysis of standardized test results, like PLANEA in Mexico. This research paper presents the analysis of academic achievement in mathematics of higher-middle education students, before and after the COVID-19 pandemic, in the state of Chihuahua, using an approach based on computational intelligence. To reach the goal, an unsupervised learning approach based on K-means was used, an algorithm that models a set of input variables and assigns elements (data vectors) with similar characteristics to certain clusters. The solution method consisted of four stages: data acquisition, variable selection, algorithm application, and number of clusters definition. As a result, a differentiated segmentation was obtained for the evaluated periods (2017 and 2022). Furthermore, based on this segmentation, some characteristics were identified for each level of academic achievement reached by the students: deficient, elementary, satisfactory, and outstanding.