<p>Personality is a psychological construct that explains different patterns of human behavior through a set of stable and measurable traits. Personality serves as the foundation of psychology and plays a crucial role in applications such as authorship attribution, enhancing recommender systems, and sentiment analysis. However, due to the lack of scalability in a large statistical community, personality recognition cannot be traditionally used through individual questionnaires. With the ever-growing spread of social networks and the vast amount of data that people leave behind, the attention of various researchers has shifted to the recognition of personality traits using social media data. This study proposes a Personality Recognition approach using Transformer Networks called PRTNet that extracts the Big-5 personality traits utilizing published text data by users on social networks. For this purpose, the SBERT (Sentence BERT) model is used to extract a collection of textual features. These features are then input into two distinct deep neural networks, single-trait and multi-trait models, which are developed to recognize personality traits. The single-trait model evaluates each personality trait independently, while the multi-trait model analyzes all five traits collectively, accounting for their interdependencies. The evaluation results of both models demonstrate that PRTNet outperforms state-of-the-art methods in different criteria, including accuracy, F-measure, MAE, MSE, and RMSE, using three well-known datasets in personality recognition, Essay, MyPersonality, and PAN.</p>