Since the existence of human life, the act of storing acquired knowlwdge about facts and events has been establishing an important role regarding human development. However, for every individual, there is a unique perception of the universe they are living in. Therefore, artifacts made for knowledge storage and representation purposes, showed up in several different arrangements, influenced by many cultural, geographical and temporal factors, which is completely surmised. However, on the 21 century, the exponetial growth of technology, led the world facinga myriad of information coming from multitudinoud sources. Then, finding ways of storing knowledge committed to certain rules became imperious. Given this scenario, this work presents a brief explanation on Knowledge Organization Systems (KOSs) and how they showed up during the last centuries. An instance of a KOSs class are the onyologies, which have been playing an important role on, for exemple, making the semantics of the real world connected to data, data in which, without such ontological commitment, could be intertpreted as represatations of different entities than the one it is, leading to biased analysis and inaccurate prediction on data-driven projects. This study will, based on works showing the benefits of bringing ontologies to the scenario of Data Science, make an application of the Human Disease Ontology, so enrichment on similaruty measures, between group of diseases annotated in on Human Disease Ontology (DO) will be made. The step of collecting data will be done considering the SIVEP-Gripe Data Set. Then, an analysis will be made on how better Machine Learning Algotithms can perform the analysis is made considering semantic rather than just numerical features.