Application of Large Language Models for Aspect-Based Sentiment Analysis on Social Media Data: A Case Study of the Thai Telecommunications Industry

Simple item page
dc.contributor.authorLimseesawan, Krittapas
dc.contributor.authorNetisopakul, Ponrudee
dc.contributor.authorChotipant, Supannada
dc.contributor.authorVoravuthikunchai, Winn
dc.contributor.authorSirivorachodphokin, Thitirat
dc.date.accessioned2026-08-06T10:53:24Z
dc.date.available2026-08-06T10:53:24Z
dc.date.issued2026-01-01
dc.description.abstractSocial media is a valuable source of consumer opinion data, particularly in Thailand's highly competitive telecommunications market among AIS, TRUE, and DTAC. This study compares three LLM-based approaches for Aspect-Based Sentiment Analysis (ABSA) on 3,508 Thai-language messages from Pantip.com and YouTube.com: (1) Prompt Engineering, (2) Domain Adaptation & Fine-Tuning, and (3) Multi-Agent Debate Framework. Results show that Domain Adaptation & Fine-Tuning achieves the best accuracy-latency trade-off, with Qwen2.5-1.5B-Instruct exceeding 87% average accuracy in under 4 seconds per message. Multi-Agent Debate achieves the highest Category accuracy (82.82%) but at a latency cost of 10-16 seconds per message. Gemini-2.5-Flash-Lite provides the best speed-accuracy balance for Prompt Engineering without additional training. Critically, small open-source models (1B-1.5B parameters) subjected to domain-specific fine-tuning can approach or surpass proprietary large models on this task, suggesting domain alignment may outweigh raw parameter scale for narrow, well-defined ABSA tasks.
dc.identifier.citationProceedings 23rd International Joint Conference on Computer Science and Software Engineering Jcsse 2026, 817-822, 2026
dc.identifier.doi10.1109/JCSSE68839.2026.11596913
dc.identifier.other2-s2.0-105045249176
dc.identifier.urihttps://dspace.kmitl.ac.th/handle/123456789/17543
dc.sourceProceedings 23rd International Joint Conference on Computer Science and Software Engineering Jcsse 2026
dc.subjectAspect-Based Sentiment Analysis
dc.subjectDomain Adaptation
dc.subjectFine-Tuning
dc.subjectLarge Language Models
dc.subjectMulti-Agent Debate
dc.subjectPrompt Engineering
dc.titleApplication of Large Language Models for Aspect-Based Sentiment Analysis on Social Media Data: A Case Study of the Thai Telecommunications Industry
dc.typeConference Paper

Files

Collections

Research