Enhancing Textual Personality Detection toward Social Media: Integrating Long-term and Short-term Perspectives

2024-04-23

arXiv.org

AI Summary

Existing research on textual personality detection typically models either long-term traits or short-term states in isolation, failing to holistically characterize personality. This work addresses the challenge by proposing the first long-term–short-term personality co-modeling paradigm for social media text, introducing the Dual Enhanced Network (DEN). DEN comprises a long-term personality encoder, a short-term personality encoder, and a bidirectional interaction module, enabling deep integration of stable trait representations with dynamic state representations. By transcending the limitations of single-temporal-view modeling, DEN achieves end-to-end joint learning. Evaluated on two benchmark personality detection datasets, DEN significantly outperforms state-of-the-art methods in both classification accuracy and psychological interpretability, empirically validating the critical value of co-modeling long- and short-term personality dimensions for comprehensive personality understanding.

Technology Category

• Cognitive Modeling & Cognitive Systems: Social Cognition And Interaction
• Machine Learning: Life-Long and Continual Learning
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• User Modeling, Personalization and Recommendation: Social recommender systems and personalization for the social Web
• Web Mining and Content Analysis: Models for Web evolution
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems

Abstract

Textual personality detection aims to identify personality characteristics by analyzing user-generated content toward social media platforms. Numerous psychological literature highlighted that personality encompasses both long-term stable traits and short-term dynamic states. However, existing studies often concentrate only on either long-term or short-term personality representations, without effectively combining both aspects. This limitation hinders a comprehensive understanding of individuals' personalities, as both stable traits and dynamic states are vital. To bridge this gap, we propose a Dual Enhanced Network(DEN) to jointly model users' long-term and short-term personality for textual personality detection. In DEN, a Long-term Personality Encoding is devised to effectively model long-term stable personality traits. Short-term Personality Encoding is presented to capture short-term dynamic personality states. The Bi-directional Interaction component facilitates the integration of both personality aspects, allowing for a comprehensive representation of the user's personality. Experimental results on two personality detection datasets demonstrate the effectiveness of the DEN model and the benefits of considering both the dynamic and stable nature of personality characteristics for textual personality detection.

Problem

• Integrating long-term stable and short-term dynamic personality traits
• Overcoming limitations of existing single-perspective personality detection methods
• Creating comprehensive personality representations from social media content

Innovation

• Dual Enhanced Network models long-term and short-term personality traits
• Long-term module analyzes consistent psychological entity patterns
• Short-term module captures contextual post information dynamically