Sentiment Analysis

Follow this topic Ask about this topic

Sentiment analysis studies how people express polarity, emotion, intensity, and opinions about specific targets, supporting applications from customer feedback and public discourse to health and finance. Common approaches range from lexicons and classical classifiers to fine-tuned transformers and large language models; aspect-based, dimensional, and structured methods capture more than a single positive-or-negative label. Multimodal systems combine words with speech, prosody, facial expression, or video, while multilingual and domain-adaptive methods aim to handle differences in language, dialect, and context. Strong benchmark results coexist with persistent challenges in neutral and implicit sentiment, class imbalance, cross-domain transfer, model bias, and disagreement over what labels actually measure; prompted LLMs do not consistently outperform task-specific fine-tuning.

Latest papers

127 papers16 papers in the last four weeks, up 167% on the four weeks before. 0.1% of all new papers.

Extracting Consumer Insight from Text: A Large Language Model Approach to Emotion and Evaluation Measurement
Feb 17, 2026Stephan Ludwig, Peter J. Danaher, Xiaohao Yang +4Sentiment Analysis LLM Information Extraction
Monash University, Australia · University of New South Wales, Australia · Flinders University, Australia +3
destroR: A Benchmark and Adversarial-Training Defense for Bangla Transfer Models under Meaning-Preserving Attacks
Nov 13, 2025Saadat Rafid Ahmed, Rubayet Shareen, Radoan Sharkar +3Adversarial Text Attacks Adversarial Robustness of Language Models
Dept. of Computer Science and Engineering, BRAC University
Polarity Detection of Sustainable Development Goals in News Text
Sep 24, 2025Andrea Cadeddu, Alessandro Chessa, Vincenzo De Leo +5LLM Evaluation Sentiment Analysis
Linkalab s.r.l., Viale Elmas, 142, Cagliari, 09122, Italy · Department of Mathematics and Computer Science, University of Cagliari, via Ospedale2026 72, Cagliari, 09124, Italy · Knowledge Media Institute, The Open University, Walton Hall, Kents Hill, Milton Keynes, MK76AA, United Kingdom
Resource-Limited Joint Multimodal Sentiment Reasoning and Classification via Chain-of-Thought Enhancement and Distillation
Aug 7, 2025Haonan Shangguan, Xiaocui Yang, Shi Feng +3Efficient Multimodal Inference Sentiment Analysis
School of Computer Science and Engineering, Northeastern University, Shenyang, China
Disentangling Bias by Modeling Intra- and Inter-modal Causal Attention for Multimodal Sentiment Analysis
Aug 7, 2025Menghua Jiang, Yuxia Lin, Baoliang Chen +3Multimodal Robustness Sentiment Analysis
School of Computer Science, South China Normal University, Guangzhou, China, 510631 · School of Electronics and Information Technology, Sun Yat-sen University, Guangzhou, China, 510275
The Super Emotion Dataset
May 21, 2025Enric Junqué de FortunyEmotion Recognition Sentiment Analysis
Managerial Decision Making IESE, Barcelona, Spain
Multi-Label Sentiment Analysis on 100 Languages with Dynamic Weighting for Label Imbalance
Aug 26, 2020Selim F. Yilmaz, E. Batuhan Kaynak, Aykut Koç +2Multilingual Sentiment Analysis Sentiment Analysis
Department of Electrical and Electronics Engineering, Bilkent University, 06800 Ankara, Turkey · Department of Computer Engineering, Bilkent University, 06800 Ankara, Turkey · National Magnetic Resonance Center UMRAM, 06800 Ankara, Turkey +1

Feb 17, 2026J/K move · Enter open · S save

Extracting Consumer Insight from Text: A Large Language Model Approach to Emotion and Evaluation Measurement

Stephan Ludwig, Peter J. Danaher, Xiaohao Yang, Yu-Ting Lin+3

Monash University, Australia · University of New South Wales, Australia · Flinders University, Australia+3

Sentiment Analysis LLM Information Extraction

Open paper AskPDF arXiv

Accurately measuring consumer emotions and evaluations from unstructured text remains a core challenge for marketing research and practice. This study introduces the Linguistic eXtractor (LX), a fine-tuned, large language model trained on consumer-authored text that also has been labeled with consumers' self-reported ratings of 16 consumption-related emotions and four evaluation constructs: trust, commitment, recommendation, and sentiment. LX consistently outperforms leading models, including GPT-4 Turbo, RoBERTa, and DeepSeek, achieving 81% macro-F1 accuracy on open-ended survey responses and greater than 95% accuracy on third-party-annotated Amazon and Yelp reviews. An application of LX to online retail data, using seemingly unrelated regression, affirms that review-expressed emotions predict product ratings, which in turn predict purchase behavior. Most emotional effects are mediated by product ratings, though some emotions, such as discontent and peacefulness, influence purchase directly, indicating that emotional tone provides meaningful signals beyond star ratings. To support its use, a no-code, cost-free, LX web application is available, enabling scalable analyses of consumer-authored text. In establishing a new methodological foundation for consumer perception measurement, this research demonstrates new methods for leveraging large language models to advance marketing research and practice, thereby achieving validated detection of marketing constructs from consumer data.