Improve Medicine Prescribing Performance Using Recommendation Systems

In this study, we introduce a decision support system for prescribing medication based on the application of association rules and analysis of prescription history data. Additionally, the research proposes techniques for analyzing prescription data, constructing a storage database, selecting rule sets for information extraction, data visualization, and improving the recommendation system’s effectiveness. The experimental results on a real data set, which includes 76,028 health insurance prescriptions, 363,697 data lines, 71 prescribing doctors, 521 medications, and 1,129 disease codes according to the ICD (International Statistical Classification of Diseases and Related Health Problems). Experimental results show that the Apriori algorithm gives good results with high accuracy, improves the efficiency and accuracy of drug prescriptions, support for doctors in making more accurate prescribing decisions and reducing patient risk.