Personality Inference from Text Using Natural Language Processing: An Exploration of Machine Learning and Deep Learning Approaches
摘要
Personality classification refers to the process of allocating distinct personality categories to humans based on their characteristics and behaviors. This review paper examines the landscape of personality prediction by analyzing text samples through natural language processing (NLP). Surveying an array of data sources, from text used on social media platforms to conventional psychological questionnaire responses, we intend to explore the different techniques ranging from different text representation methods, open and closed vocabulary approaches, machine learning algorithms and deep learning architectures employed in decoding the relation between language patterns in text and personality traits. The key contributions of this survey include an overview of recent approaches in personality prediction from text, a comparative analysis of techniques and models across datasets, and the identification of future research directions, including a combination of demographic factors, exploring multidimensional analysis, and using psychological questionnaires.