HyperPersona: A hypergraph framework for text-based personality prediction
In brief
- HyperPersona models text hierarchy (document, sentence, word) as a hypergraph structure.
- A transformer-based graph encoder learns interactions across linguistic layers.
- The model outperforms existing approaches in predicting the Big Five personality traits.
HyperPersona introduces a novel framework for text-based personality prediction by modeling text’s hierarchical structure (document, sentence, word) as a hypergraph. The model combines this representation with a transformer-based graph encoder to capture multi-level dependencies. Results on the Big Five personality traits show significant improvements over existing methods.
This summary gives you the essentials. The full article, its figures and quotes are on cs.AI updates on arXiv.org.
Read the original articleAI-generated summary of the source, checked by rules. Report an error