Abstract
Type 2 diabetes is a chronic disease that is caused by a combination of genetic and environmental factors, with social determinants playing a significant role. These social determinants include but not limited to factors such as education level, occupation, family history, ethnicity and place of residence etc. The prevalence of Type 2 diabetes (T2D) necessitates inventive management strategies. This study explores the potential of utilizing Ripple Down Rules (RDR) in comparison with Machine Learning (ML) approaches for the incremental development of a Knowledge-Based System (KBS). The aim is to create a robust Decision Support System (DSS) for effective T2D management, focusing on the dynamic influence of evolving social determinants. The research outcomes substantiate the viability of this approach. The KBS, fortified by RDRs, demonstrated notable performance metrics. Specifically, the system achieved an impressive accuracy rate of 90.2%, accompanied by a specificity of 96.9%. The sensitivity, crucial for identifying potential T2D cases, reached 73.6%, indicating the system's proficiency in recognizing diverse instances.
Recommended Citation
Omar, Adel; Beydoun, Ghassan; Win, Khin Than; Shukla, Nagesh; Jelinek, Herbert; and Elias, Hector, "Cultivating Expertise: Unravelling Type 2 Diabetes Associations through Incremental Knowledge-Based System Development: Ripple Down Rules or Machine Learning" (2023). ACIS 2023 Proceedings. 96.
https://aisel.aisnet.org/acis2023/96