HyperPersona: A Multi-Level Hypergraph Framework for Text-Based Automatic Personality Prediction

2026-05-17

AI Summary

This work addresses the limitation of existing automatic personality prediction methods, which often overlook the multi-level hierarchical structure of language—spanning documents, sentences, and words—and consequently struggle to effectively integrate global, local, and fine-grained semantic information. To overcome this, the authors propose HyperPersona, a novel framework that introduces hypergraphs into personality prediction for the first time, explicitly modeling the tripartite linguistic hierarchy by treating words as nodes and sentences and documents as hyperedges. A Transformer-based graph encoder is designed to jointly capture intra-level interactions and inter-level dependencies, yielding context-sensitive and structure-aware personality representations. Using only raw text as input, HyperPersona significantly outperforms current state-of-the-art models across all Big Five personality dimensions, demonstrating the critical role of multi-level textual structure in personality inference.

Abstract

As a modern commodity, language has become a vast repository of socially and psychologically significant traits and concepts, reflecting the ways people encode pattern of thoughts, behaviors, and emotions into words. Text-based Automatic Personality Prediction (APP), seeks to infer personality from linguistic behavior, offering a scalable alternative to traditional psychometric assessments. Although text is inherently hierarchical, with the document-level capturing global features, the sentence-level encoding local semantics, and the word-level providing fine-grained lexical information, most existing approaches rely on shallow, sequential, or single-level representations that ignore the multi-level structure of written language. To address this, we propose HyperPersona, a framework that explicitly models the hierarchical organization of text (document, sentence, and word) through hypergraph structure, where a document and its sentences are represented as hyperedges, and the words are represented as nodes, enabling joint modeling of global, local, and lexical dependencies of text. Followed by a transformer-based graph encoder that learns interactions within and across these linguistic layers, yielding context-sensitive and structurally grounded feature representations for personality prediction. Experiments on the Big Five personality dimensions show that, while relying solely on text, HyperPersona effectively integrates multi-level linguistic cues, achieving superior performance compared to state-of-the-art baselines. These findings underscore the critical role of textual hierarchy in advancing human-like personality inference from natural language.

Problem

• Automatic Personality Prediction
• text hierarchy
• multi-level representation
• linguistic structure
• personality inference

Innovation

• hypergraph
• hierarchical text modeling
• automatic personality prediction
• multi-level representation
• transformer-based graph encoder