Unveiling Personality Profiles: A Data-Driven Approach to Big Five Traits, Sentiment, and Textual Self-Description

Unveiling Personality Profiles: A Data-Driven Approach to Big Five Traits, Sentiment, and Textual Self-Description

Abstract

This study presents a data-driven framework for understanding personality traits, emotional expression, and self-description using Natural Language Processing and psychometric analysis. It integrates the Big Five Inventory, sentiment analysis, semantic embeddings, clustering, and regression modelling to investigate behavioural and emotional patterns in textual self-expression. Data was collected through survey responses containing personality-inventory items, subjective-happiness indicators, lifestyle-satisfaction measures, and open-ended textual responses. Sentence Transformers, emotion classification, zero-shot classification, K-Means clustering, and UMAP dimensionality reduction were used in the analysis. The findings demonstrate significant associations between personality traits and subjective well-being while highlighting the effectiveness of textual analysis in behavioural profiling.

Authors

Krishna Pratap Rao, Ramanand Roy, Rampravesh Kumar, Nishant Kumar, Nisha Kumari

Institution

Noida Institute of Engineering & Technology (MCA Institute), Greater Noida, India

PDF Link

View Full Paper