Machine Learning and Deep Learning Approaches to Predict Early Stage of Diabetes

Abstract

This study offers a thorough methodology that makes use of a variety of deep learning and machine learning algorithms to predict early stage of diabetes. Recurrent neural networks (RNN), Feedforward Neural Networks (FNN), Decision Trees (DT), Logistic Regression (LR), Random Forests (RF), K-Nearest Neighbors (KNN), Naive Bayes (NB), Support Vector Machines (SVM), and Long Short-Term Memory (LSTM) are all integrated in the suggested system. Nine health attributes for 100,000 entries are included in the dataset, which was obtained via Kaggle. Exploratory data analysis, quality checks, and encoding are all part of data pre-processing. For model evaluation, the dataset is divided into training and test sets, and a two-pronged feature selection technique is used. Notably, with 97% accuracy, the Decision Tree machine learning model shows greater accuracy in diabetes prediction. The study places a strong emphasis on moral issues with predictive modeling in healthcare. Prospective avenues for investigation encompass improving prediction models, augmenting openness, and tackling wider ethical considerations in the field of healthcare analytics.