Ruixian Yang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Ruixian Yang has a total of 16 co-authors. Here are the top co-authors by number of shared publications:
- Yan Jin
School of information management, Zhengzhou University, Zhengzhou, China
3 shared publications - Zhuo Sun
School of Information Management, Zhengzhou University
3 shared publications - Chongwu Bi
School of Information Management, Zhengzhou University
2 shared publications Chongwu Bi
Zhengzhou University
1 shared publication- Di Zhao
Zhengzhou University
1 shared publication Jiwen Zhang
Zhengzhou University
1 shared publication
Latest Publications
- Open Access
- 20 March 2026
- Published by University of Borås
- Vol. 31 (iConf) , 178-192
Limitations in cultural context: systematic biases of LLMs in Chinese sentiment analysis
Introduction. Large language models (LLMs) are increasingly applied in Chinese sentiment analysis; however, their ability to interpret sentiment in culturally specific contexts remains uncertain. This study systematically evaluates the potential biases of LLM-based Chinese sentiment analysis, providing an empirical basis for model optimisation and responsible application. Method. A human-AI comparison experiment was conducted at both word and sentence levels. Two hundred Chinese words from four categories and 225 sentences were used as materials. Judgment data on sentiment polarity, intensity, valence, and arousal were collected from native Chinese speakers and representative Chinese and international LLMs. Data analysis was performed using Chi-square tests and Kruskal–Wallis tests. Results. In sentiment polarity judgment, LLMs are highly consistent with humans and outperform the traditional sentiment lexicon. For continuous dimensions (intensity, valence, arousal), LLMs generally show an exaggeration bias, especially for sensory-perceptual words and ironic sentences. The bias stems from the disembodied cognitive nature of models, exaggerated linguistic patterns in training data, and technical characteristics such as the attention mechanism. Conclusion. This study reveals a systematic limitation of LLMs in Chinese sentiment analysis, characterised by accurate classification but exaggerated quantitative evaluations. Consequently, in practical applications, outputs related to sentiment intensity should be interpreted with caution.
- 20 March 2026
- Published by University of Borås
- Vol. 31 (iConf) , 178-192
Open AccessLimitations in cultural context: systematic biases of LLMs in Chinese sentiment analysis
Introduction. Large language models (LLMs) are increasingly applied in Chinese sentiment analysis; however, their ability to interpret sentiment in culturally specific contexts remains uncertain. This study systematically evaluates the potential biases of LLM-based Chinese sentiment analysis, providing an empirical basis for model optimisation and responsible application. Method. A human-AI comparison experiment was conducted at both word and sentence levels. Two hundred Chinese words from four categories and 225 sentences were used as materials. Judgment data on sentiment polarity, intensity, valence, and arousal were collected from native Chinese speakers and representative Chinese and international LLMs. Data analysis was performed using Chi-square tests and Kruskal–Wallis tests. Results. In sentiment polarity judgment, LLMs are highly consistent with humans and outperform the traditional sentiment lexicon. For continuous dimensions (intensity, valence, arousal), LLMs generally show an exaggeration bias, especially for sensory-perceptual words and ironic sentences. The bias stems from the disembodied cognitive nature of models, exaggerated linguistic patterns in training data, and technical characteristics such as the attention mechanism. Conclusion. This study reveals a systematic limitation of LLMs in Chinese sentiment analysis, characterised by accurate classification but exaggerated quantitative evaluations. Consequently, in practical applications, outputs related to sentiment intensity should be interpreted with caution.
- 1 September 2025
- Published by Springer Nature
- p. 264-272
Privacy Risks in the Storytelling of Open Government Data: A Study from the Perspective of User Cognitive Reasoning
- 1 September 2025
- Published by Springer Nature
- p. 264-272
Privacy Risks in the Storytelling of Open Government Data: A Study from the Perspective of User Cognitive Reasoning
- Open Access
- 16 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-3
Investigating the Determinants of User Adoption Behavior in the Context of Conflicting Health Information: a DEMATEL-ISM Model Approach
- 16 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-3
Open AccessInvestigating the Determinants of User Adoption Behavior in the Context of Conflicting Health Information: a DEMATEL-ISM Model Approach
- Open Access
- 16 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-3
Knowledge Graph Construction of Chinese Traditional Yu Opera Based on Joint Entity-Relation Extraction Method
- 16 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-3
Open AccessKnowledge Graph Construction of Chinese Traditional Yu Opera Based on Joint Entity-Relation Extraction Method