The research employs Naive Bayes and Support Vector Machine (SVM) classification techniques to analyze attitudes toward cigarette consumption based on Twitter user opinions. Twitter, being one of the most popular social media platforms, serves as an excellent source for gauging public sentiment on various issues, including cigarette smoking, referred to here as "X." The diverse array of opinions poses a challenge for accurate sentiment classification. This study evaluates the effectiveness of the Naive Bayes and SVM algorithms in categorizing sentiment as positive, negative, or neutral. Data is collected through web scraping, and preprocessing steps such as text cleaning, tokenization, and stemming are implemented. The performance of the classification is assessed using metrics like accuracy, precision, recall, and F1-score. The results indicate that SVM outperforms Naive Bayes in sentiment analysis related to cigarette use. These findings provide new insights into public opinion and aim to assist policymakers in developing effective tobacco control strategies.