Exploring Scale-Measures of Data Sets
This work provides a new algorithm for efficiently exploring scale-measures, which could benefit researchers and practitioners in data-driven science by enabling semi-automatic data scaling.
This paper explores the properties of the lattice of scale-measures for data sets, which is a theoretical framework for scaling data. The authors propose a novel scale-measure exploration algorithm, based on attribute exploration, that enables efficient exploration of these scale-measures.
Measurement is a fundamental building block of numerous scientific models and their creation. This is in particular true for data driven science. Due to the high complexity and size of modern data sets, the necessity for the development of understandable and efficient scaling methods is at hand. A profound theory for scaling data is scale-measures, as developed in the field of formal concept analysis. Recent developments indicate that the set of all scale-measures for a given data set constitutes a lattice and does hence allow efficient exploring algorithms. In this work we study the properties of said lattice and propose a novel scale-measure exploration algorithm that is based on the well-known and proven attribute exploration approach. Our results motivate multiple applications in scale recommendation, most prominently (semi-)automatic scaling.