Supervised learning in general is affected by two key problems: high dimensionality of the input data and limited number of labeled samples. The high dimensionality problem can be addressed through feature reduction strategies. Active learning (AL), which aims to select the most informative samples for training the model, has also been demonstrated to be an effective approach for dealing with the limited availability of labeled samples. While most research has focused on addressing the two aforementioned problems independently, I aim to integrate the feature reduction and AL steps into a unique framework to solve the two fundamental supervised learning problems together.