Visualising Computational Intelligence Through Converting Data into Formal Concepts
Simon Andrews, Constantinos Orphanides, Simon M. Polovina · 2010
Formal Concept Analysis (FCA) is an emerging data technology that complements collective intelligence such as that identified in the Semantic Web by visualising the hidden meaning in disparate and distributed data. The paper demonstrates the discovery of these novel semantics through a set of FCA open source software tools FcaBedrock and In-Close that were developed by the authors. These tools add computational intelligence by converting data into a Boolean form called a Formal Context, prepare this data for analysis by creating focused and noise-free sub-Contexts and then analyse the prepared data using a visualisation called a Concept Lattice. The Formal Concepts thus visualised highlight how data itself contains meaning, and how FCA tools thereby extract data's inherent semantics. The paper describes how this will be further developed in a project called CUBIST, to provide in-data-warehouse visual analytics for RDF-based triple stores.