Data Mining continues its battle against dense volumes of data poured in, to withstand the urge for knowledge discovery. Knowledge enhances the decision-making capability providing a peak edge in the competitive world. The medical domain is one such application, where a fast and wise decision gains to be a lifesaver. Several mining techniques already prevailed in the medical world with their essence reflected in their advancements. Among the discrete usage scenarios, Drug similarity prediction is evolving to be an attractive choice of research, as new drug development is expensive and time-consuming. Also, the approval rate of the FDA is stepping down, which paves attention towards drug repositioning based on similarities and a quench for optimized results. Our research addresses this for the mentioned complications by employing a hybrid approach. Our framework blends the features of KNN with ACO to attain enhanced Drug consumption similarities. The proposed hybrid computational method promises leveraged results, simulated with JSIM evaluating performance parameters like recall, accuracy, and time.