AUTOMATIC PERSONALITY DETECTION USING DEEP LEARNING MODELS IN USER-GENERATED CONTENT

Abstract

This thesis titled "Automatic Personality Detection Using Deep Learning Models in User-Generated Content" presents a cutting-edge exploration into the realm of artificial intelligence, focusing on the application of deep learning models to discern and analyze personality traits from vast arrays of user-generated content (UGC). This work stands at the intersection of computer science, psychology, and data analytics, endeavouring to bridge the gap between quantitative AI techniques and qualitative psychological insights. Central to the research is the utilization of advanced machine learning models, particularly transformer-based architectures like BERT and GPT, renowned for their ability to process and understand natural language at a granular level. These models are applied to a variety of data sources, including social media posts, blog entries, and online interactions, to extract and classify personality indicators according to well-established frameworks such as the Big Five personality traits and the Myers-Briggs Type Indicator (MBTI). The research is innovative in its approach, using not just textual analysis but also exploring multimodal data, including video and audio, to capture the nuances of personality expression that text alone might miss. By integrating deep learning with psychological theories, the study aims to develop models that not only predict personality types with high accuracy but also understand the underlying human behaviours and emotional states that drive online interactions. One of the significant contributions of this research is its focus on cross linguistic and cross-cultural applicability. Recognizing the diversity of global online communities, the study tests and validates models across different languages and cultural contexts, enhancing the generalizability and usefulness of the findings. This aspect is crucial, considering the potential for AI systems to perpetuate biases if not properly trained on diverse datasets. The thesis also explores the practical application of its findings in real-life domain, demonstrating how personality detection can be utilized to improve recruitment processes by aligning candidates' personalities with organizational culture. In detailing the results, the research discusses the performance of various models tested, providing comparative analyses that highlight the strengths and limitations of each approach. In conclusion, this thesis not only advances the field of AI and personality psychology but also contributes to the broader discourse on the integration of technology and human-centric sciences. By forging a link between computational models and psychological theories, it lays a foundation for future innovations that respect human diversity and foster more personalized and meaningful interactions in the digital age.