Development of an AI-Driven Model for Drug Sales Prediction Using Enhanced Golden Eagle Optimization and XGBoost Algorithm

Abstract

The pharmaceutical industry plays a crucial role in public health by providing essential medications to address various medical conditions. Predicting drug sales accurately is paramount for pharmaceutical companies to efficiently manage their resources, plan production, and optimize marketing strategies. To address this an AI-based model for predicting medication product sales based on the Enhanced Golden Eagle Optimized and Extreme Gradient Boosting (EGEO-XGBoost) framework. The technique begins with data collecting from the Kaggle website, followed by pre-processing with Min-max normalization to remove noise and assure consistency. The preprocessed data is then used to extract relevant features via Linear Discriminant Analysis (LDA). The enhanced EGEO method fine-tunes the parameters of the XGBoost model, improving its predictive ability. The comparative findings demonstrate that the proposed method significantly improves accuracy (0.90), specificity (0.86), sensitivity (0.92), MCC (0.82), F1-score (0.94), and RMSE (4.02). Incorporating such predictive models into the decision-making processes of pharmaceutical corporations can result in improved resource management, better marketing plans, and increased operational effectiveness.