In the modern era, there has been a substantial increase in health-related concerns among individuals. The prevalence of various diseases has surged significantly, underscoring the crucial necessity for timely disease detection. A strategy employed for this purpose involves the examination of drug evaluations from diverse patient groups. Within the realm of healthcare, reviews assume a pivotal role. They offer insights into the efficacy of different drugs and their potential adverse effects. The primary objective is to categorize a patient’s health condition based on their feedback pertaining to a specific medication. In this endeavor, a system for classifying medical conditions has been developed, utilizing patient reviews related to diverse medications. This system has the ability to forecast the specific ailment afflicting a patient. To achieve this objective, the Drug Review Dataset from UCI’s Machine Learning Repository has been employed. To extract relevant features, multiple vectorization techniques have been applied, including Bag of Words, TF-IDF, TF-IDF (Bigrams), and TF-IDF (Trigrams). To execute the necessary categorization, a range of machine learning classification algorithms have been harnessed, such as Decision Tree, Multinomial Naive Bayes, Logistic Regression, and Passive Aggressive Classifier.