Some useful representations can be derived. This is not done directly from the data; it is achieved by using machine learning procedures, which inform us about how the algorithms operate and offer us a more precise overview of the role of each predictor in the predictions obtained. In particular, learning curves can provide a quick diagnosis to improve your models. It helps you figure out whether you need more observations, or you need to enrich your variables.
A learning curve is a useful diagnostic graphic that depicts the behavior of your machine learning algorithm (your hypothesis) with respect to the available quantity of observations. The idea is to compare how the training performance (the error or accuracy of the in-sample cases) behaves with respect to the cross-validation (usually ten-fold) using different in-sample sizes.