Women’s participation in Computing remains marked by structural inequalities that intensify when race, territory, and other social dimensions are considered together. Although the literature discusses intersectionality in this field, most research remains theoretical or relies solely on descriptive statistics, with a scarcity of empirical studies that employ quantitative methods to identify intersectional profiles that account for the overlap of intersectional dimensions. This study applies clustering techniques to institutional data on students enrolled in Computing courses at a Brazilian public university, adopting a two-step approach: (i) clustering of the general population and (ii) clustering restricted to women students. The results show that aggregate analysis is insufficient to reveal minority female profiles due to the severe population imbalance, which reproduces dominant patterns. In contrast, clustering restricted to women identifies three distinct intersectional clusters, marked especially by differences in race/ethnicity and academic pathways. These findings highlight the potential of quantitative data mining methods to reveal intersectional dynamics that are often invisible in traditional analyses, providing support for equity policies that are more sensitive to the specificities of women students in Computing.