FUZZY MAPREDUCE BASED INTUITIVE RANDOM FOREST MODEL FOR DIABETIC MELLITUS DETECTION USING BIG DATA

Abstract

The exponential growth of healthcare data, particularly in the context of chronic diseases like diabetes mellitus, necessitates the development of scalable and accurate analytical models. This paper proposes a novel hybrid approach, termed Fuzzy MapReduce based Intuitive Random Forest Model (FMIRFM), for effective detection of diabetes mellitus from large-scale datasets. The proposed architecture integrates two key functional modules: the Exclusive Fuzzy MapReduce Module, which facilitates optimized parallel processing by constructing fast execution threads and distributing them across multiple servers, and the Knowledgebase-rooted Random Forest Diabetes Predictor, which enhances classification accuracy by leveraging domain-specific patterns. The FMIRFM method ensures robust performance even with high-dimensional and voluminous data, improving both processing speed and diagnostic accuracy. Comprehensive experimental evaluation demonstrates that the proposed model significantly outperforms traditional approaches in terms of accuracy, precision, sensitivity, specificity, and F-score. These results affirm the capability of the FMIRFM framework to serve as an effective diagnostic aid in large-scale clinical environments.