Through the process of individualizing the selection of drug and dose based on an individual's genetic profile, personalized medicine gives patients access to treatment that is both safer and more effective. The response of cancer patients to anti-cancer treatments (drugs) is one of the most significant issues in personalized medicine, which is important because it allows for the release of the target treatment. Researchers have been encouraged to construct Artificial Intelligence (AI) based models for predicting medication response in order to advance cancer treatment. This motivation stems from the magnitude and availability of data regarding drug sensitivity respectively. Among the artificial intelligence models that are of concern are Machine Learning (ML) and Deep Learning (DL) based models, which have recently evolved. The purpose of this research is to present two methods for predicting medication response: a data federation approach and a DL-based model. The primary objective is to generalize the predictor in order to ensure that it is capable of reliably predicting the reaction to a variety of medications. In order to consolidate the data, the data federation is applied. This is due to the fact that the data has a significant impact on any AI model. When it comes to using AI algorithms to solve challenges in personalized medicine, such as disease detection or prediction, accurate disease diagnosis, and therapy optimization, the choice of the algorithm, which is impacted by its capacity and applicability, is an important consideration. The purpose of this research is to examine the applicability and capability of artificial neural networks (ANN), support vector machines (SVM), Naïve Bayes, and fuzzy logic in the context of solving problems related to personalized medicine. The findings produced are found to be satisfactory and in line with the expectations.