This study proposes a machine learning (ML)-based e-diagnostic system for the early detection of gestational diabetes mellitus (GDM). The research explores the application of the K-nearest neighbors (KNN) algorithm to predict diabetes diagnosis using the Pima Indians Diabetes Database. The KNN algorithm, a non-parametric, instance-based learning method, was used to classify individuals as diabetic or non-diabetic. The study aimed to evaluate the algorithm's ability to make accurate predictions and explore factors influencing its performance. Data preprocessing steps included handling missing values, feature scaling, and splitting the data into training and testing sets. The KNN classifier was trained and tested with these parameters, resulting in an accuracy of approximately 0.76 in predicting diabetes diagnosis. The study also discusses the significance of hyperparameter tuning, data preprocessing, and handling imbalanced data for optimal KNN performance. The findings suggest that KNN can serve as a viable tool for early diabetes detection, with potential applications in healthcare and predictive modeling. The study also reviews existing literature on machine learning approaches for diabetes classification, including recall, accuracy, precision, and F1-score. The research highlights the importance of feature selection and the use of KNN for early diabetes detection. The study concludes that KNN is a promising method for diabetes prediction, with further research needed to improve its performance. The study also discusses the importance of data preprocessing, feature selection, and model evaluation in achieving accurate diabetes prediction. The results show that KNN can be used to predict diabetes with an accuracy of 76.56%. The study also discusses the importance of cross-validation and test/train split in evaluating the performance of the KNN model. The study concludes that KNN is a suitable method for diabetes prediction, with potential applications in healthcare and predictive modeling. The study also discusses the importance of data preprocessing, feature selection, and model evaluation in achieving accurate diabetes prediction. The study concludes that KNN is a suitable method for diabetes prediction, with potential applications in healthcare and predictive modeling.This study proposes a machine learning (ML)-based e-diagnostic system for the early detection of gestational diabetes mellitus (GDM). The research explores the application of the K-nearest neighbors (KNN) algorithm to predict diabetes diagnosis using the Pima Indians Diabetes Database. The KNN algorithm, a non-parametric, instance-based learning method, was used to classify individuals as diabetic or non-diabetic. The study aimed to evaluate the algorithm's ability to make accurate predictions and explore factors influencing its performance. Data preprocessing steps included handling missing values, feature scaling, and splitting the data into training and testing sets. The KNN classifier was trained and tested with these parameters, resulting in an accuracy of approximately 0.76 in predicting diabetes diagnosis. The study also discusses the significance of hyperparameter tuning, data preprocessing, and handling imbalanced data for optimal KNN performance. The findings suggest that KNN can serve as a viable tool for early diabetes detection, with potential applications in healthcare and predictive modeling. The study also reviews existing literature on machine learning approaches for diabetes classification, including recall, accuracy, precision, and F1-score. The research highlights the importance of feature selection and the use of KNN for early diabetes detection. The study concludes that KNN is a promising method for diabetes prediction, with further research needed to improve its performance. The study also discusses the importance of data preprocessing, feature selection, and model evaluation in achieving accurate diabetes prediction. The results show that KNN can be used to predict diabetes with an accuracy of 76.56%. The study also discusses the importance of cross-validation and test/train split in evaluating the performance of the KNN model. The study concludes that KNN is a suitable method for diabetes prediction, with potential applications in healthcare and predictive modeling. The study also discusses the importance of data preprocessing, feature selection, and model evaluation in achieving accurate diabetes prediction. The study concludes that KNN is a suitable method for diabetes prediction, with potential applications in healthcare and predictive modeling.