Data Mining Framework for Discovering and Clustering Phenotypes of Atypical Diabetic Lung

The global diabetes epidemic has exposed significant clinical variability in diabetic patients, including phenotypes that defy accepted characteristics. Atypical diabetic phenotypes in lungs are sifting through numerous data sets to establish connections and identify patterns to use data analysis to solve a certain issue. This article presents the Data Mining Framework based on the Atypical Diabetes Prediction Model (DMFADPM), built on the Data Mining Framework to identify and group the characteristics of A.D. in lungs. Data Mining Framework is used to solve the main problems of improving the prediction model's accuracy and enabling the model's adaptation to multiple datasets. The exclusion and inclusion processes are the two components of the preprocessing steps that form the foundation of the data mining technique. The experimental analysis of the proposed method has obtained an accuracy of 98.9%, prediction ratio of 0.22%, gain ratio of 95%, information gain of 92%.