Drug Classification and Repurposing Using Decision Tree Algorithm and Data Analysis

Abstract

Abstract: Fundamentally, drugs are substances that exhibit the capacity to alter biological functions, often serving as interventions to mitigate ailments or enhance physiological processes. However, the landscape of drug development and usage is undergoing transformation, fueled by innovative strategies that hold the promise of revolutionizing healthcare solutions. This research paper explores drug classification and repurposing through the utilization of decision tree algorithms and data analysis. Specifically, the aim is to classify drugs into three distinct types: Drug X which is most used drugs, Drug Y which is less frequently used drugs, and Drug C includes drugs like narcotics. Decision tree algorithms are employed to discern the defining attributes that categorize drugs into these types. By analyzing comprehensive datasets encompassing drug properties, interactions, and clinical outcomes, the study harnesses decision tree models to predict drug classifications accurately. This approach holds the potential to accelerate drug discovery, optimize treatment strategies, and contribute to more efficient healthcare solutions. The proposed algorithm is compared with both Naïve Bayes and K - Nearest Neighbors algorithms to prove it is more accurate. Ultimately, the significance of this paper transcends the boundaries of conventional drug development paradigms and to overcome the problem of shortage of drugs.