Abstract:
"Diabetes's increasing prevalence needs the development of robust predicting technologies as
well as tailored therapies. This study uses a machine learning framework to tackle the complex
job of detecting diabetes type and developing individualized food regimens for individual
diabetic patients. The vital need for precise diabetes type prediction and personalized dietary
recommendations is emphasized.
The study uses the Random Forest method for its robustness and incorporates important
characteristics such as gender, age, hypertension, heart disease, smoking history, BMI, HbA1c
level, and glucose levels into the entire feature set. Model training entails meticulously splitting
the dataset into discrete training and testing sets to guarantee the proposed approach's
dependability.
Preliminary data are reported using quantitative indicators such as the confusion matrix,
classification report, and accuracy score. These measures provide subtle insights into the
model's performance, demonstrating its ability to appropriately categorize various diabetes
kinds. The provided accuracy score serves as an initial benchmark, indicating the model's
ability to predict diabetes kinds using the supplied variables."
