Advancements in Personality Detection: Unleashing the Power of Transformer-Based Models and Deep Learning with Static Embeddings on English Personality Quotes

Personality detection has garnered significant attention in recent years, with applications ranging from personalized user experiences to psychological analysis. This paper presents advancements in personality detection, focusing on the utilization of Transformer-based models and deep learning models with static embeddings to analyse English personality quotes. The research highlights the capabilities of advanced models such as ELECTRA and META OPT in comprehending contextual dependencies within text. Concurrently, it examines the significance of deep learning and embeddings in capturing semantic information and hidden personality traits. Leveraging the power of modern natural language processing techniques, the study explores the potential of these models in extracting latent personality traits from textual data. A diverse dataset of English quotes with personality dimension along the introversion-extroversion spectrum, supplemented by the concept of ambiverts is curated for training and evaluation, and the model's performance is assessed using accuracy, precision, recall and F1-score. The results reveal that the Transformer-based models significantly enhances personality detection accuracy compared to conventional methods. By exploiting these advanced techniques, the research contributes to a deeper understanding of individual personalities through their textual expressions, bridging the gap between human cognition and artificial intelligence to revolutionize personalized interactions.