Growing demands on paper quality impel papermakers to found new, robust, and reliable methodologies to evaluate paper properties, directly aligned with consumer preferences. In this paper, we address the development of such methodolgies, focused on two different aspects of the paper surface: surface roughness and surface waviness. Using profilometry measurements and quality assessments made by a panel of experts, we have developed and tested several classification methodologies in order to emulate their evaluations, by combining different feature extraction techniques with classifiers representing it variety of statistical methods commonly applied. The current standard method to infer roughness, the Bendtsen tester, was also analyzed, in order to cast sortie light on how its measurements are affected by fundamental parameters of the paper surface. Results obtained show that it is indeed possible to develop a suitable classifier for inferring the quality of paper roughness in three classes (good/medium/bad) and for paper waviness in two classes (good/not good).