Abstract
The apparent reliance on data and experimental evidence in system modeling, decision-making, pattern recognition, and control engineering entails their centrality and a paramount role of data science. To capture the essence of data and avoid various artifacts (noise, outliers, incompleteness) as well as facilitate building essential descriptors and revealing key relationships, we advocate a need of transforming data into information granules. Information granules are regarded as conceptually sound knowledge tidbits over which a number of various models are developed.