Data clustering is recently a common techniqueto group similar data with certain features. It enables find-ing the representative in each cluster as well. However, theclustering analysis comprises several challenging tasks, e.g.,feature selection, choice among different clustering algorithms,defining the optimal cluster number, clustering with the use ofa distance measure dealing with various levels of measurement,cluster validation, and interpretation of results in the end.The objective of this paper is the conceptual design of ascenario catalog including extracted representative near-crashand crash scenarios. Two clustering algorithms based on k-covers and k-medoids are applied to data in a naturalisticdriving study under consideration of aforementioned aspects.Afterwards, the comparison of two clustering algorithms isconducted based on the cluster representativeness, purity, andaverage silhouette width. Moreover, the clusters are visualizedin a two dimensional scenario space by t-Distributed StochasticNeighbor Embedding (t-SNE). The derived scenario catalogcovers the selected database at best possible rate and enablesa cost-efficient development of predictive safety functions.