Abstract
The aim of this research paper is to explore the potential of machine learning techniques in predicting the utilization of the Electronic Prescription Service (EPS) and Electronic Repeat Dispensing (eRD) items to categorize General Practitioner (GP) practices based on their usage patterns. The study utilized raw data related to dispensaries, EPS, and eRD acquired from the National Health Service online medical database. To achieve this objective, exploratory data analysis was conducted on the dataset, which was then split into a training set and a testing set. Various machine learning algorithms, including linear regression, decision tree regression, and random forest regression, were applied to the training set to develop a predictive model. The models were evaluated using measurements such as the “Score”, “Mean Squared Error (MSE)”, “Mean Absolute Error (MAE)”, “Sqrt Mean Absolute Error (MAE)” and “Coefficient of determination (R^2)”. The study found that the machine learning models developed were effective in predicting EPS utilisation and could categorize GP practices based on their usage patterns. This categorization could help identify high-utilization practices, leading to more efficient resource allocation and ultimately improved healthcare delivery. The results also indicate the potential for machine learning techniques to predict the utilization of other healthcare services and could pave the way for more personalized and targeted healthcare services in the future.