Skip to content
Conference Open access

HLLC: A Four-Stage Human-AI Collaborative Annotation Framework for Panic Emotion Recognition in Social Networks

Aug 2026 · 2026 12th International Conference on Big Data and Information Analytics (BigDIA) · pp. 290-297 · 0 citations · 26 references

TL;DR

This paper leverages large language models (LLMs) and proposes a four-stage progressive human-AI collaborative annotation framework named HLLC (Human-LLM Collaborative Labeling), enabling large-scale, fine-grained panic emotion dataset construction at low cost.

Abstract

Public emergencies often trigger large-scale panic emotions, which are further amplified by social media, potentially leading to serious consequences such as resource hoarding and social disorder. Therefore, timely identification and intervention of panic are of great significance. However, the development of panic emotion recognition models is severely constrained by the scarcity of high-quality annotated datasets. Existing coarse-grained sentiment classification fails to distinguish panic from fear and anxiety, where the latter two are characterized by high arousal and a sense of loss of control. Furthermore, purely manual annotation is expensive and hard to scale, while fully automated annotation suffers from model hallucinations and conservative bias. To address the challenges in constructing large-scale panic emotion datasets, this paper leverages large language models (LLMs) and proposes a four-stage progressive human-AI collaborative annotation framework named HLLC (Human-LLM Collaborative Labeling). The framework sequentially performs data sampling, LLM initial labeling, crowdsourced verification, and model generalization. Through representative sampling, multi-model consistency comparison for zero-shot labeling, confidence-based sample routing, and BERT fine-tuning, the framework balances annotation quality and efficiency. Experiments on the CrisisNLP Hurricane Sandy dataset [1] show that the HLLC framework achieves a panic F1 score of 0.787 and an overall accuracy of 0.857, improving by 33.3% over lexicon methods and boosting panic recall by 78.7% over pure LLM annotation. These results demonstrate that HLLC enables large-scale, fine-grained panic emotion dataset construction at low cost, offering a practical solution for emergency public opinion monitoring and early warning.

Read PDF

Similar papers

Open access Oct 2026

Development of methods for extracting situations, emotional assessments and values from social media texts

This paper presents an end-to-end pipeline for automated social media discourse analysis, integrating situation extraction from natural language texts, emotion classification, and value identification. Most existing approaches are limited to binary sentiment analysis and do not account for the value context underlying...

Ko Hein, D. Palchunov · 0 citations
#natural language process... Preprint Oct 2026

PAT: A Personality-Driven Augmentation and Transfer Learning Framework for Depression Detection on Social Media

Depression is a prevalent mental disorder that affects millions worldwide, with early detection crucial for effective intervention. While exist?ing methods have achieved promising results in automated depression detection, they face two key limitations: (1) noise in post screening: symptom-based semantic screening is e...

Shen Zheng, Yu-Zhe Ding, Xiao-Hui Liang et al. · 0 citations
#small language model Book Open access Nov 2026

A Long-sequence Hierarchical Multimodal Fusion Framework for Personality-aware Depression Detection

Depression detection in older adults is an important human-aware multimedia problem, where latent mental-health states must be inferred from non-invasive behavioral and individual-difference cues. This task is challenging because depressive symptoms are often subtle, heterogeneous, and reflected across multiple channel...

Bi-Chen Wang, Yu-Fei Hou, Yong-Hui Ji et al. · 0 citations
Oct 2026

Enhancing emotion recognition using lexical-contextual fusion with attention-based DistilBERT

A modified DistilBERT that incorporates lexical emotion representations into the transformer pipeline via a feature fusion layer with attention weighting is proposed, which enables the joint learning of semantic dependencies, lexical cues, and text embeddings, which enhances the recognition of underrepresented emotions...

Manjog Padhy, Rasmita Rautray · 0 citations
Open access Oct 2026

Corpus linguistic-driven hybrid feature representation learning with graph convolutional network for text-based emotion detection

This research introduces a Hybrid Attention Learning with Graph Convolutional Network for Text-Based Emotion Detection (HALGCN-TED) framework, aiming to develop a robust and automated model to improve the representation and detection of complex emotional patterns.

Mohammad A. Mahzari, Noah N. Alsager, A. Hilal · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.