Data Trading Similarity Signature An Extended Data Trading Framework for Human and Non-Human Actors

Abstract

Fair and secure data trading is one of the most prominent challenges of the 21st century. This paper presents a second iteration of an approach to develop a data marketplace concept by checking consumer requirements. The main problem we identified is data quality and the question: Would a dataset fulfill the consumer requirements? Starting from an approach that uses a binary response set to answer the question of whether requirements are met, we concluded that a description of consumer requirements needs to be quantitatively comparable. The novel approach presented here identifies similarities between datasets and consumer requirements. It forms a unique, fingerprint-like similarity signature for each dataset, which can be interpreted by both human and non-human actors. The approach is deducted and designed by using the Design Science Research Methodology and discussed critically in the end.