In today’s digital environment, the use of social media platforms is constantly increasing, leading to a substantial influx of textual content and multimedia such as images being generated and shared online on a daily basis. Leveraging data obtained from social media for personality prediction presents a promising avenue, primarily due to its ability to circumvent the requirement of users filling out extensive questionnaires. This approach not only saves time but also enhances the credibility of the results obtained. In the scope of our ongoing research, we have actively collected real-time data from Twitter using a widely accepted application programming interface, which has allowed us to compile a preliminary dataset for experimental purposes. Employing a variety of well-established Machine Learning algorithms, like k-Nearest Neighbor, Naive Bayes, and Support Vector Machine, we sought to explore the effectiveness of these methods in predicting personality traits based on social media data. Notably, our investigation has revealed that the Naive Bayes algorithm demonstrated the most promising results, exhibiting an accuracy rate of 71.67%. This outcome signifies that, among the classifiers employed, Naive Bayes proved that it works most effectively in accurately classifying personality traits based on the collected social media data.