Diabetes has dramatically increased the risk of various cardiovascular problems and previous approaches employed for diabetic prediction using machine learning algorithms are laced with curse of dimensionality and poor prediction accuracy issues. As a result, this research employs Recurrent Neural Network model, a subtype of deep learning to build a model for the prediction of Early-stage diabetes in patients. The dataset for training the model was obtained from Kaggle and a hospital in Nigeria. The dataset was preprocessed by removing duplicate and irrelevant entries, removing outliers, and handling missing data. Results from evaluation show that the diabetes prediction model performed well with an accuracy of 0.90, precision of 0.91, Recall of 0.95 and F1-Score of 0.93. The developed model outperformed conventional machine learning techniques and it has the potential to help healthcare professionals anticipate and prevent diabetes.
