Using Psycholinguistic Clues to Index Deep Semantic Evidences - Personality Detection in Social Media Texts

Sep 17, 2026

Uploaded by Fans

Personality detection in social media content targets dominant personality traits from short, noisy posts, moving beyond earlier approaches that used a single coherent writing sample. Existing attention and graph-based methods struggle to build meaningful connections because posts are fragmented and disjointed, especially under limited supervision. The study proposes an index attention mechanism guided by psycholinguistic prior knowledge and introduces the Index Attention Network (IAN) for targeted deep-semantic fusion, achieving strong results on Kaggle and competitive performance on Pandora.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages15

Reading Time38 min

AccessFree to download

Abstract

Introduction

Language use and personality

Progression of Personality Detection in Texts

RESEARCH ARTICLE Using Psycholinguistic Clues to Index Deep Semantic Evidences: Personality Detection in Social Media Texts Qirui Tang1, Wenkang Jiang2, Xinlong Pan3,*, Lei Lin4, Jizhao Zhu5, Yihua Du4,* and Donghong Sun6 1 School of Cyber Security, University of Chinese Academy of Sciences, Beijing 101408, China 2 Australian Institute for Machine Learning, The University of Adelaide, Adelaide 5005, Australia 3 Institute of Information Fusion, Naval Aviation University, Yantai 264001, China 4 Computer Network Information Center, Chinese Academy of Sciences, Beijing 100085, China 5 School of Computer Science, Shenyang Aerospace University, Shenyang 110136, China 6 Institute for Network Science and Cyberspace, Tsinghua University, Beijing 100084, China Abstract Detecting personalities in social media content isan important application of personality psychology. Most early studies apply a coherent piece of writing to personality detection, but today, the challenge is to identify dominant personality traits from a series of short, noisy social media posts. To this end, recent studies have attempted to individually encode the deep semantics of posts, often using attention-based methods, and then relate them, or directly assemble them into graph structures. However, due to the inherently disjointed and noisy nature of social media content, constructing meaningful connections remains challenging. While such methods rely on well-deﬁned relationships between posts, eﬀectively capturing these connections in fragmented Academic Editor: Yulong Huang Submitted: 04 March 2025 Accepted: 15 April 2025 Published: 26 April 2025 Vol. 2, No. 2, 2025 . 10.62762/CJIF.2025.820998 *Corresponding authors: 􀀀 Xinlong Pan [airadar@126.com](airadar@126.com) 􀀀 Yihua Du [yhdu@cashq.ac.cn](yhdu@cashq.ac.cn) and sparse content is non-trivial, particularly under limited supervision or noisy input. To tackle this, we draw inspiration from the scanning reading technique—commonly recommended for eﬃciently processing large volumes of information—and propose an index attention mechanism as a solution. This mechanism leverages prior psycholinguistic knowledge as an “index” to guide attention, thereby enabling moreeﬀective information fusion across scattered semantic signals. Building on this idea, we introduce the Index Attention Network (IAN)—a novel framework designed to infer personality labels by performing targeted information fusion over deep semantic representations of individual posts. Through a series of experiments, IAN achieved state-of-the-art performance on the Kaggle dataset and performance comparable to graph convolutional networks (GCN) on the Pandora dataset. Notably, IAN delivered an average improvement of 13% in terms of macro-F1 scores with the Kaggle dataset. The code for IAN is available at GitHub: [https://github.com/Once2gain/IAN](https://github.com/Once2gain/IAN). Keywords: personality detection, attention mechanism, Citation Tang, Q., Jiang, W., Pan, X., Lin, L., Zhu, J., Du, Y., & Sun, D. (2025) . Using Psycholinguistic Clues to Index Deep Semantic Evidences: Personality Detection in Social Media Texts. Chinese Journal of Information Fusion, 2(2), 112–126 . © 2025 by the Authors. Published by Institute of Central Computation and Knowledge. This is an open access article under the CC BY license ([https://creati](https://creati)[vecommons.org/licenses/by/4.0/](vecommons.org/licenses/by/4.0/)) . social media text mining, information fusion. 1 Introduction Personality testing is a common task in psychology. The traditional method of testing a subject’s personality is an artiﬁcially designed questionnaire, which is relatively reliable but not particularly eﬃcient. To this end, researchers have proposed several diﬀerent automated methods of testing personalities. Of these, analyzing user-generated content is one of the most important [1, 2] . Fortunately, social media provides vas

FAQs

Why is personality detection on social media more challenging than traditional personality testing?

It must infer dominant traits from multiple short, noisy posts rather than from a single controlled questionnaire or coherent writing sample, making semantic connections harder to construct.

What limitation do existing attention or graph-based methods face on social media text?

They rely on well-defined relationships between posts, but social media content is inherently disjointed and sparse, so capturing useful connections is difficult under limited supervision.

How does the proposed Index Attention Network (IAN) address this problem?

IAN uses an index attention mechanism guided by prior psycholinguistic knowledge to fuse scattered deep semantic signals and infer personality labels via targeted information fusion.

Sep 17, 2026

Uploaded by Fans