Can Large Language Models be Effective Online Opinion Miners?
Ryang Heo, Yongsik Seo, Junseong Lee, Dongha Lee
Abstract
The surge of user-generated online content presents a wealth of insights into customer preferences and market trends. However, the highly diverse, complex, and context-rich nature of such content poses significant challenges to traditional opinion mining approaches. To address this, we introduce Online Opinion Mining Benchmark (OOMB), a novel dataset and evaluation protocol designed to assess the ability of large language models (LLMs) to mine opinions effectively from diverse and intricate online environments. OOMB provides, for each content instance, an extensive set of (entity, feature, opinion) tuples and a corresponding opinion-centric insight that highlights key opinion topics, thereby enabling the evaluation of both the extractive and abstractive capabilities of models. Through our proposed benchmark, we conduct a comprehensive analysis of which aspects remain challenging and where LLMs exhibit adaptability, to explore whether they can effectively serve as opinion miners in realistic online scenarios. This study lays the foundation for LLM-based opinion mining and discusses directions for future research in this field. Our code and dataset are available 1 . † Corresponding author 1 https://github.com/ryang1119/OOMB I read an article on the 2024 Hyundai Tucson, and it highlights the improved interior design. The new layout seems more .. The 2024 Tesla Model 3 offers an incredibly smooth ride with its advanced suspension system. < User-Generated Online Data Collection > 2025 Kia EV9: 4 reasons to love it, 2 reasons to think twice Blog Reason to love it #1: Styling If you're looking for something boxy that oozes "SUV" while actually being a family hauler in disguise, the Kia EV9 has your number. … Reason to think twice it #1: Ho-Hum Interior design I've been driving the 2025 Ford Explorer for a few months now, and it's been an excellent family SUV. The interior is spacious, with .. 2025 Ford Explorer A Well-Rounded Family SUV ⭐⭐⭐⭐⭐ (5/5) Review Site 2025 Chevrolet Traverse Spacious but Lacks Refinement ⭐⭐⭐ (3/5) I recently purchased the 2025 Chevrolet Traverse, it has some great qualities, but there are a few disappointments as well. First, .. where t p = (e p , f p , o p ) and t g = (e g , f g , o g ) are the predicted and gold tuples, respectively. Drawing on recent works (Han et al., 2023a; Li et al., 2024) , we utilize the Python's difflib library 4 to 4 https://docs.python.org/3/library/difflib.html
Ask about this paper
Your agent reads all of it.
Lune indexed this paper to the last equation, along with the top-tier papers that cite it. Ask a question and the answer quotes them.
Cited by top-tier papers1
Ask how each one uses itBuilds on6
- BERTScore: Evaluating Text Generation with BERTTianyi Zhang, Varsha Kishore, Felix Wu, Kilian Q. Weinberger et al.ICLR 2020 · 8,443 citations
- Aspect Sentiment Quad Prediction as Paraphrase GenerationWenxuan Zhang, Yang Deng, Xin Li, Yifei Yuan et al.EMNLP 2021 · 196 citations
- SpanMlt: A Span-based Multi-Task Learning Framework for Pair-wise Aspect and Opinion Terms ExtractionHe Zhao, Longtao Huang, Rong Zhang, Quan Lu et al.ACL 2020 · 174 citations
- Large Language Models for Data Annotation and Synthesis: A SurveyZhen Tan, Dawei Li, Song Wang, Alimohammad Beigi et al.EMNLP 2024 · 119 citations
- Deep Weighted MaxSAT for Aspect-based Opinion ExtractionMeixi Wu, Wenya Wang, Sinno Jialin PanEMNLP 2020 · 25 citations
Related papers
- MindVote: When AI Meets the Wild West of Social Media OpinionXutao Mao, Ezra Xuanru Tao, Leyao WangAAAI 2026
- Beyond Binary: Towards Fine-Grained LLM-Generated Text Detection via Role Recognition and Involvement MeasurementZihao Cheng, Li Zhou, Feng Jiang, Benyou Wang et al.WWW 2025 · 20 citations
- EEmo-Bench: A Benchmark for Multi-modal Large Language Models on Image Evoked Emotion AssessmentLancheng Gao, Ziheng Jia, Yunhao Zeng, Wei Sun et al.ACM MM 2025 · 2 citations
- AIR-Bench: Automated Heterogeneous Information Retrieval BenchmarkJianlyu Chen, Nan Wang, Chaofan Li, Bo Wang et al.ACL 2025
- OpenTuringBench: An Open-Model-based Benchmark and Framework for Machine-Generated Text Detection and AttributionLucio La Cava, Andrea TagarelliEMNLP 2025
