A Controlled Natural Language to Support Intent-based Blockchain Selection

Eder John Scheid, Patrick Widmer, Bruno Rodrigues, Muriel Franco, Burkhard Stiller

2nd IEEE International Conference on Blockchain and Cryptocurrency (ICBC 2020), Toronto, Canada, May 2020, pp 1–9

Abstract

In the last years, cryptocurrencies have become increasingly popular along with their underlying distributed ledger technology, referred to as a Blockchain (BC). Nowadays, a wide variety of BC implementations are available. However, the selection of a suitable implementation for a particular application or use case is complex because it requires technical understanding of the underlying BC implementation aspects. Therefore, this paper proposes a Controlled Natural Language (CNL) to extends existing BC selection solutions to abstract underlying implementation details. The approach allows the specification abstract high-level policies, referred to as intents, in an English-based language. The approach is inspired by previous approaches from the network management field. Moreover, a state machine-based refinement technique is proposed to refine these intents into low-level BC selection policies. The results of the performance evaluation of the prototype implementation show that the refinement process presents a minimal overhead. In addition, the perceived intuitiveness of the CNL by users was assessed in a survey. The results of the survey suggest that technical and non-technical individuals benefit from an intentbased approach equally.