OLAP navigation in the Granular Linguistic Model of a Phenomenon

The amount of data provided by computers about objects in our environment increases. Nevertheless, the ability to extract relevant and understandable knowledge from this information remains limited. On-Line Analytical Processing (OLAP) is a well known paradigm used to help users to navigate by the information stored in databases. In the research line of Computational Theory of Perceptions, we have created the Granular Linguistic Model of a Phenomenon. It is a data structure that allows computational systems to generate linguistic descriptions of input data. In this paper, we explore the possibilities of using OLAP to navigate in this structure of information. Inspired in the way humans use NL, we adapt the typical operations in OLAP, namely, drilling, rolling and slicing/dicing, to navigate by hierarchical granular structures of fuzzy perceptions. The long term aim is to create a new type of human computer interface that will assist users in the analysis of the likely huge amount of available information about relevant phenomena. Obtained results show the viability of this approach including a practical demonstration of concept.