Sentiment analysis has become an essential tool in data mining and text classification, enabling the extraction of subjective information from diverse sources. This study analyzes student feedback on 19 central universities participating in CUCET 2022, as featured on the educational website Shiksha.com. The primary objectives of this study were to evaluate student perceptions of these universities and rank them based on the proportion of positive reviews. Furthermore, we assessed the performance of two machine learning classifiers, Naive Bayes and support vector machine (SVM), in predicting review sentiments. Our results indicated that Jawaharlal Nehru University, India, garnered the highest percentage of positive reviews, while Assam University received the lowest. The SVM model exhibited superior predictive accuracy (75%) compared to the Naive Bayes model (56%), underscoring its robustness in sentiment prediction for this dataset. Nevertheless, the study faced limitations, including the varying number of reviews per university and data quality issues, which could impact the accuracy and rankings. To further demonstrate the versatility of sentiment analysis, we conducted a case study on IMDb movie reviews using Python. This involved data preprocessing, building a sentiment analysis model with TextBlob, and evaluating its performance, which emphasized the broader applicability and effectiveness of sentiment analysis tools across different contexts.