|
|
基于情感增强机制的大语言模型虚假新闻检测
|
Abstract:
为解决现有新闻文本虚假检测方法仅依赖语义特征、忽视情感特征，导致复杂内容检测准确度低的问题，提出一种基于情感增强机制的大语言模型虚假新闻检测方法(Sentiment-Enhanced Large Language Model for Fake News Detection, SELLM-FND)。该方法先对新闻文本进行情感分析以提取情感特征，再通过大语言模型融合文本与情感特征完成检测。在WELFake_Dataset_Edited数据集上的实验显示，该方法准确率达0.929，检测性能优于以往基于文本的虚假新闻检测方法。
In order to solve the problem that the existing false news detection methods only rely on semantic features and ignore emotional features, which leads to the low accuracy of complex content detection, a sentient-enhanced large language model for false news detection (SELLM-FND) based on emotional enhancement mechanism is proposed. This method firstly analyzes the news text to extract emotional features, and then completes the detection by fusing the text and emotional features through the large language model. Experiments on WELFake_Dataset_Edited data set show that the accuracy of this method is 0.929, and the detection performance is better than the previous text-based false news detection methods.
| [1] | Shu, K., Sliva, A., Wang, S., Tang, J. and Liu, H. (2017) Fake News Detection on Social Media: A Data Mining Perspective. ACM SIGKDD Explorations Newsletter, 19, 22-36. https://doi.org/10.1145/3137597.3137600 |
| [2] | Capuano, N., Fenza, G., Loia, V. and Nota, F.D. (2023) Content-Based Fake News Detection with Machine and Deep Learning: A Systematic Review. Neurocomputing, 530, 91-103. https://doi.org/10.1016/j.neucom.2023.02.005 |
| [3] | Hirlekar, V.V. and Kumar, A. (2020) Natural Language Processing Based Online Fake News Detection Challenges—A Detailed Review. 2020 5th International Conference on Communication and Electronics Systems (ICCES), Coimbatore, 10-12 June 2020, 748-754. https://doi.org/10.1109/icces48766.2020.9137915 |
| [4] | Shen, Y., Liu, Q., Guo, N., Yuan, J. and Yang, Y. (2023) Fake News Detection on Social Networks: A Survey. Applied Sciences, 13, Article 11877. https://doi.org/10.3390/app132111877 |
| [5] | Phan, H.T., Nguyen, N.T. and Hwang, D. (2023) Fake News Detection: A Survey of Graph Neural Network Methods. Applied Soft Computing, 139, Article ID: 110235. https://doi.org/10.1016/j.asoc.2023.110235 |
| [6] | Ahn, Y. and Jeong, C. (2019) Natural Language Contents Evaluation System for Detecting Fake News Using Deep Learning. 2019 16th International Joint Conference on Computer Science and Software Engineering (JCSSE), Chonburi, 10-12 July 2019, 289-92. https://doi.org/10.1109/jcsse.2019.8864171 |
| [7] | Upadhayay, B. and Behzadan, V. (2020) Sentimental LIAR: Extended Corpus and Deep Learning Models for Fake Claim Classification. 2020 IEEE International Conference on Intelligence and Security Informatics (ISI), Arlington, 9-10 November 2020, 1-6. https://doi.org/10.1109/isi49825.2020.9280528 |
| [8] | Yuan, C., Qian, W., Ma, Q., Zhou, W. and Hu, S. (2021) SRLF: A Stance-Aware Reinforcement Learning Framework for Content-Based Rumor Detection on Social Media. 2021 International Joint Conference on Neural Networks (IJCNN), Shenzhen, 18-22 July 2021, 1-8. https://doi.org/10.1109/ijcnn52387.2021.9533444 |
| [9] | Zhang, X. and Ghorbani, A.A. (2020) An Overview of Online Fake News: Characterization, Detection, and Discussion. Information Processing & Management, 57, Article ID: 102025. https://doi.org/10.1016/j.ipm.2019.03.004 |
| [10] | Nasir, J.A., Khan, O.S. and Varlamis, I. (2021) Fake News Detection: A Hybrid CNN-RNN Based Deep Learning Approach. International Journal of Information Management Data Insights, 1, Article ID: 100007. https://doi.org/10.1016/j.jjimei.2020.100007 |
| [11] | Dimpas, P.K., Po, R.V. and Sabellano, M.J. (2020) Filipino and English Clickbait Detection Using a Long Short Term Memory Recurrent Neural Network. 2017 International Conference on Asian Language Processing (IALP), Singapore, 5-7 December 2017, 276-280. |
| [12] | Telang, H., More, S., Modi, Y. and Kurup, L. (2019) Anempirical Analysis of Classification Models for Detection of Fake News Articles. 2019 IEEE International Conference on Electrical, Computer and Communication Technologies (ICECCT), Coimbatore, 20-22 February 2019, 1-7. https://doi.org/10.1109/icecct.2019.8869504 |
| [13] | Farokhian, M., Rafe, V. and Veisi, H. (2023) Fake News Detection Using Dual BERT Deep Neural Networks. Multimedia Tools and Applications, 83, 43831-43848. https://doi.org/10.1007/s11042-023-17115-w |
| [14] | Hu, B., Sheng, Q., Cao, J., Shi, Y., Li, Y., Wang, D., et al. (2024) Bad Actor, Good Advisor: Exploring the Role of Large Language Models in Fake News Detection. Proceedings of the AAAI Conference on Artificial Intelligence, 38, 22105-22113. https://doi.org/10.1609/aaai.v38i20.30214 |
| [15] | Ke, J. (2024) An Implicit Semantic Enhanced Fine-Grained Fake News Detection Method Based on Large Language Models. Journal of Computer Research and Development, 61, 1250-1260. |
| [16] | Hussein, D.M.E.M. (2018) A Survey on Sentiment Analysis Challenges. Journal of King Saud University—Engineering Sciences, 30, 330-338. https://doi.org/10.1016/j.jksues.2016.04.002 |
| [17] | Kalbhor, S. and Goyal, D. (2023). Survey on ABSA Based on Machine Learning, Deep Learning and Transfer Learning Approach. AIP Conference Proceedings, 2782, Article ID: 020041. https://doi.org/10.1063/5.0154549 |
| [18] | Zhong, B. (2024) Fine-grained Sentiment Analysis Using Multidimensional Feature Fusion and GCN. Journal of Information and Telecommunication, 9, 91-112. https://doi.org/10.1080/24751839.2024.2386785 |
| [19] | Nkhata, G., Gauch, S. and Anjum, U. (2025) Fine-Tuning BERT with Bidirectional LSTM for Fine-Grained Movie Reviews Sentiment Analysis. arXiv: 2502.20682. |
| [20] | Teng, J., He, H. and HU, G. (2025) A Fine-Grained Sentiment Recognition Method for Online Government-Public Interaction Texts Based on Large Language Models. International Conference on Artificial Intelligence and Machine Learning Research (CAIMLR 2024), Singapore, 28-29 September 2024. https://doi.org/10.1117/12.3058735 |
| [21] | Gu, J.W. (2025) A Survey on LLM-As-A-Judge. arXiv: 2411.15594. |
| [22] | Boissonneault, D. and Hensen, E. (2024) Fake News Detection with Large Language Models on the LIAR Dataset. https://doi.org/10.21203/rs.3.rs-4465815/v1 |
| [23] | Han, Z.Y. (2024) Parameter-Efficient Fine-Tuning for Large Models: A Comprehensive Survey. arXiv: 2403.14608. |
| [24] | 王东清, 芦飞, 张炳会, 等. 大语言模型中提示词工程综述[J]. 计算机系统应用, 2025, 34(1): 1-10. |
| [25] | Reynolds, L. and McDonell, K. (2021) Prompt Programming for Large Language Models: Beyond the Few-Shot Paradigm. Extended Abstracts of the 2021 CHI Conference on Human Factors in Computing Systems, Yokohama, 8-13 May 2021, 1-7. https://doi.org/10.1145/3411763.3451760 |
| [26] | Kojima, T., et al. (2022) Large Language Models are Zero-Shot Reasoners. arXiv: 2205.11916. |
| [27] | Yao, S.Y., et al. (2023) ReAct: Synergizing Reasoning and Acting in Language Models. arXiv: 2210.03629. |
| [28] | Yao, S.Y., et al. (2023) Tree of Thoughts: Deliberate Problem Solving with Large Language Models. arXiv: 2305.10601. |
| [29] | Besta, M., et al. (2023) Graph of Thoughts: Solving Elaborate Problems with Large Language Models. arXiv: 2308.09687. |
| [30] | Zhou, Y.C., Muresanu, A., Han, Z.W., et al. (2023) Large Language Models Are Human-Level Prompt Engineers. International Conference on Learning Representations. https://iclr.cc/virtual/2023/10850 |
| [31] | Lewis, P., et al. (2020) Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks. arXiv: 2005.11401. |
| [32] | Etaiwi, W. and Alhijawi, B. (2025) Comparative Evaluation of ChatGPT and DeepSeek across Key NLP Tasks: Strengths, Weaknesses, and Domain-Specific Performance. Array, 27, Article ID: 100478. https://doi.org/10.1016/j.array.2025.100478 |
| [33] | Real-Jiakai (2024) Deepseek-R1-Distill-Qwen-7B-News-Classifier. Hugging Face. https://huggingface.co/real-jiakai/Deepseek-R1-Distill-Qwen-7B-News-Classifier |
| [34] | Hu, E. (2021) LoRA: Low-Rank Adaptation of Large Language Models. arXiv: 2106.09685. |
| [35] | SemEvalWorkshop (2024) Sem_eval_2018_task_1 Dataset. Hugging Face. https://huggingface.co/datasets/SemEvalWorkshop/sem_eval_2018_task_1 |
| [36] | Summitsky (2024) WELFake_Dataset_Edited Dataset. Hugging Face. https://huggingface.co/datasets/Summitsky/WELFake_Dataset_Edited |
| [37] | Kolev, V., Weiss, G. and Spanakis, G. (2022) FOREAL: Roberta Model for Fake News Detection Based on Emotions. Proceedings of the 14th International Conference on Agents and Artificial Intelligence, 3-5 February 2022, 429-440. https://doi.org/10.5220/0010873900003116 |
| [38] | Wang, Y., Gu, Z., Zhang, S., Zheng, S., Wang, T., Li, T., et al. (2025) LLM-GAN: Constructing Generative Adversarial Network through Large Language Models for Explainable Fake News Detection. ICASSP 2025—2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Hyderabad, 6-11 April 2025, 1-5. https://doi.org/10.1109/icassp49660.2025.10889048 |
