LL4G: self-supervised dynamic optimization for graph-based personality detection
LL4G: self-supervised dynamic optimization for graph-based personality detection
Graph-based personality detection constructs graph structures from textual data, particularly social media posts. Current methods often struggle with sparse or noisy data and rely on static graphs, limiting their ability to capture dynamic changes between nodes and relationships. This paper introduces LL4G, a self-supervised framework leveraging large language models (LLMs) to optimize graph neural networks (GNNs). LLMs extract rich semantic features to generate node representations and to infer explicit and implicit relationships. The graph structure adaptively adds nodes and edges based on input data, continuously optimizing itself. The GNN then uses these optimized representations for joint training on node reconstruction, edge prediction, and contrastive learning tasks. This integration of semantic and structural information generates robust personality profiles. Experimental results on Kaggle and Pandora datasets show LL4G outperforms state-of-the-art models.
cs.CL, cs.LG
Long, Yunfei
6652ac59-2950-4738-b001-5e187655b0d8
Cai, Xiaohao
de483445-45e9-4b21-a4e8-b0427fc72cee
Chen, Guanming
a5c50691-6b41-4669-b2c1-01a95d1be450
Wang, Yuhan
6d047f0f-fa73-4ed0-8eb4-ca05a3474450
Razzak, Imran
85c57ead-8a63-4aec-bba3-559a43dd5888
Jameel, Shoaib
ae3c588e-4a59-43d9-af41-ea30d7caaf96
2 April 2025
Long, Yunfei
6652ac59-2950-4738-b001-5e187655b0d8
Cai, Xiaohao
de483445-45e9-4b21-a4e8-b0427fc72cee
Chen, Guanming
a5c50691-6b41-4669-b2c1-01a95d1be450
Wang, Yuhan
6d047f0f-fa73-4ed0-8eb4-ca05a3474450
Razzak, Imran
85c57ead-8a63-4aec-bba3-559a43dd5888
Jameel, Shoaib
ae3c588e-4a59-43d9-af41-ea30d7caaf96
[Unknown type: UNSPECIFIED]
Abstract
Graph-based personality detection constructs graph structures from textual data, particularly social media posts. Current methods often struggle with sparse or noisy data and rely on static graphs, limiting their ability to capture dynamic changes between nodes and relationships. This paper introduces LL4G, a self-supervised framework leveraging large language models (LLMs) to optimize graph neural networks (GNNs). LLMs extract rich semantic features to generate node representations and to infer explicit and implicit relationships. The graph structure adaptively adds nodes and edges based on input data, continuously optimizing itself. The GNN then uses these optimized representations for joint training on node reconstruction, edge prediction, and contrastive learning tasks. This integration of semantic and structural information generates robust personality profiles. Experimental results on Kaggle and Pandora datasets show LL4G outperforms state-of-the-art models.
Text
2504.02146v1
- Author's Original
More information
Published date: 2 April 2025
Keywords:
cs.CL, cs.LG
Identifiers
Local EPrints ID: 502150
URI: http://eprints.soton.ac.uk/id/eprint/502150
PURE UUID: 42be8ebc-49bd-461a-928a-ab18b23e633e
Catalogue record
Date deposited: 17 Jun 2025 16:48
Last modified: 18 Jun 2025 02:04
Export record
Altmetrics
Contributors
Author:
Lingzhi Shen
Author:
Yunfei Long
Author:
Xiaohao Cai
Author:
Guanming Chen
Author:
Yuhan Wang
Author:
Imran Razzak
Author:
Shoaib Jameel
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics