Controlled Natural Languages for Specifying Business Intelligence Applications

TL;DR

This paper explores the use of controlled natural languages to specify business intelligence applications, demonstrating their effectiveness through a healthcare example for improved data and interface description.

Contribution

It introduces and applies two controlled natural languages, CNL-BI and ASL, for specifying complex BI applications in a healthcare context.

Findings

CNLs effectively describe complex BI data and functions

Application of CNLs improves clarity in BI requirement specifications

Demonstrated with a healthcare-focused BI application example

Abstract

This study examines the use of controlled natural languages (CNLs) to specify business intelligence (BI) application requirements. Two varieties of CNLs, CNL-BI and ITLingo ASL (ASL), were employed. A hypothetical BI application, MEDBuddy-BI, was developed for the National Health Service (NHS) to demonstrate how the languages can be used. MEDBuddy-BI leverages patient data, including interactions and appointments, to improve healthcare services. The research outlines the application of CNL-BI and ASL in BI. It details how these languages effectively describe complex data, user interfaces, and various BI application functions. Using the MEDBuddy-BI running example.