Evaluation of classification algorithms framework domain-specific language: The case of finance-accounting domain.

Abstract

This short paper presents preliminary results on the evaluation of a Classification Algorithms Framework Domain-Specific Language (CAF DSL) that was previously developed. In our previous work, a domain-specific language to provide a framework for choosing-evaluating classification algorithms but with a system-level and simplified interface approach was presented. However, the evaluation that took place was restricted to data scientists and language engineers. In this work, the evaluation of the same DSL but by finance-accountant professionals (domain-experts) took place in a small scale. The DSL was presented and feedback regarding its usability and learnability took place. Future plans consist of scaling this attempt further and extending it to different domains such as health care. The long-term plan consists of bringing AI - Classification algorithms to professionals and to the wider public.