The field of Multiple Criteria Decision Aiding studies decision problems where multiple points of view need to be considered and aims at providing tools and methods for aiding the process of reaching these decisions. Three classical decision problems can be identified: choice, sorting and ranking. Many approaches have been developed for solving these problems since the founding of the field more than 50 years ago. Most of the studies problems, however, deal only with a handful of objects, or decision alternatives. We consider that large amounts of data related to a decision problem would be available in certain real-life problems and will use a case study to illustrate this. We then propose to extend several clustering approaches, oriented towards objects defined by data on which preferential information can be expressed, as a means of exploring such large datasets. Furthermore, the results from such an analysis may be used in a process of eliciting the parameters of the preference model used.