Abstract: The present invention relates to a customized nose pad recommendation system and method, which classifies a user's nose type based on facial scan data and recommends a nose pad suitable for the user according to the classified nose type. More specifically, the invention pertains to a technique for quantitatively classifying the user's nose by analyzing various facial indicator data related to nose shape, and for providing an optimal nose pad based on representative values and adjustment values corresponding to each nose type class. The invention calculates class-wise probabilities for each variable using Cumulative Distribution Function (CDF)-based relative distance computation, determines the class to which the user most likely belongs based on the highest probability, and refers to the corresponding adjustment value data to recommend a nose pad. The recommendation result can be iteratively adjusted and validated based on the Vertex Distance (VD) value.