基于复杂网络和情感计算的文化舆情演化模型

Cultural Public Opinion Evolution Model Based on Complex Networks and Affective Computing

摘要

针对文化舆情在网络传播中如何准确定量揭示并建模其网络结构特征与情感演化规律的问题,提出一种基于复杂网络和情感计算融合的文化舆情动态演化模型.该模型以抖音平台数据为核心,通过构建多维加权关系矩阵实现舆情传播网络的定量化表达,设计情感加权社群聚类算法识别情感共鸣,获取网络中情绪极性与传播结构的融合特征,并引入结合注意力机制的门控循环单元网络进行演化的时序预测模型,实现对情绪波动和传播趋势的动态建模.实验结果表明,该模型有效刻画了吉林文化舆情在网络空间中的传播结构和情感分布特征,其拟合精度和预测结果均优于传统HK(Hegselmann-Krause)意见动力学模型与SIR(susceptible,infected,recovered)传播模型,验证了该模型具有较强的泛化能力,可为文化的数字化传播规律分析提供有效的理论支撑.

Abstract

Aiming at the problems that how to accurately and quantitatively reveal and model the network structure characteristics and sentiment evolution patterns of cultural public opinion in online communication, we proposed a dynamic evolution model for cultural public opinion based on the integration of complex network and affective computing. The model took the data from the Douyin platform as the core, realized quantitative expression of the public opinion communication network by constructing a multi-dimensional weighted relationship matrix, designed a sentiment-weighted community clustering algorithm to identify sentiment resonance, obtained the fused features of emotional polarity and communication structure within the network, and introduced a temporal prediction model of evolution of a gated recurrent unit network combined with an attention mechanism to achieve dynamic modeling of sentiment fluctuations and communication trends. Experimental results show that the proposed model effectively characterizes the communication structure and sentiment distribution characteristics of Jilin cultural public opinion in cyberspace, and its fitting accuracy and predictive results are superior to those of traditional Hegselmann-Krause (HK) opinion dynamics models and susceptible, infected, recovered (SIR) communication models, verifying that the model has strong generalization capability and can provide effective theoretical support for analyzing the digital dissemination patterns of culture.

关键词

文化舆情 / 复杂网络 / 情感计算 / 文化舆情演化

Key words

cultural public opinion / complex network / affective computing / cultural public opinion evolution

参考文献

[1] 张向阳. 基于复杂网络的在线评论情感倾向性分类[D]. 大连: 大连理工大学, 2016.

[2] (Zhang X Y. Sentiment Classification of Online Reviews Based on Complex Networks[D]. Dalian: Dalian University of Technology, 2016.)

[3] 俞奇. 基于主题与情感的网络舆情演化模型构建及实证研究[D]. 北京: 首都经济贸易大学, 2022.

[4] (Yu Q. Construction and Empirical Study of Online Public Opinion Evolution Model Based on Topic and Sentiment[D]. Beijing: Capital University of Economics and Business, 2022.)

[5] 余传明, 周丹. 情感词汇共现网络的复杂网络特性分析[J]. 情报学报, 2010, 29(5):906-914.

[6] (Yu C M, Zhou D. Analysis of Complex Network Characteristics of Sentiment Word Co-occurrence Network[J]. Journal of the China Society for Scientific and Technical Information, 2010, 29(5): 906-914.)

[7] 刘娇, 崔荣一, 赵亚慧. 基于自联想记忆与卷积神经网络的跨语言情感分类[J]. 中文信息学报, 2018, 32(12):118-124.

[8] (Liu J, Cui R Y, Zhao Y H. Cross-Lingual Sentiment Classification Based on Auto-associative Memory and Convolutional Neural Network[J]. Journal of Chinese Information Processing, 2018, 32(12): 118-124.)

[9] 郝玫, 王道平. 中文网络评论的复杂语义倾向性计算方法研究[J]. 图书情报工作, 2014, 58(22):105-110.

[10] (Hao M, Wang D P. Research on Calculation Method of Complex Semantic Orientation for Chinese Online Reviews[J]. Library and Information Service, 2014, 58(22): 105-110.)

[11] 张华平, 李恒训, 李清敏. 情感词发现与极性权重自动计算算法研究[J]. 中文信息学报, 2017, 31(3):48-54.

[12] (Zhang H P, Li H X, Li Q M. Research on Algorithm of Sentiment Word Discovery and Polarity Weight Automatic Calculation[J]. Journal of Chinese Information Processing, 2017, 31(3): 48-54.)

[13] 关慧, 韩志远. MGE-BERT:融合标签排序与情感图感知的多标签情感分类模型[J]. 计算机系统应用, 2025, 34(3):268-276.

[14] (Guan H, Han Z Y. MGE-BERT: Multi-label Sentiment Classification Model Fusing Label Ranking and Sentiment Graph Perception[J]. Computer Systems & Applications, 2025, 34(3): 268-276.)

[15] 蔡谞诚. 融合通道注意力机制与BERT预训练模型的乏信息文本情感分析[D]. 南昌: 江西师范大学, 2024.

[16] (Cai X C. Sentiment Analysis of Poor-Information Text Fusing Channel Attention Mechanism and BERT Pre-trained Model[D]. Nanchang: Jiangxi Normal University, 2024.)

[17] 李丽, 李平. 基于交互图神经网络的方面级多模态情感分析[J]. 计算机应用研究, 2023, 40(12):3683-3689.

[18] (Li L, Li P. Aspect-Level Multimodal Sentiment Analysis Based on Interactive Graph Neural Network[J]. Application Research of Computers, 2023, 40(12): 3683-3689.)

[19] Hu C, Zhang J R, Wei X F, et al. Study of the Clinical Distinctions of Acupuncture-Moxibustion Treatment of Acute Gouty Arthritis Based on Complex Networks[J]. Journal of Acupuncture and Tuina Science, 2024, 22(3): 253-262.

[20] Jiang T H. Assessment Method of Emergency Preparedness System Vulnerability Based on the Complex Network Theory[J]. Journal of Risk Analysis and Crisis Response, 2012, 2(3): 195-200.

[21] 向宁. 佛教互联网舆情观点演化机制的行动者中心模型研究[J]. 世界宗教文化, 2017(5):143-149.

[22] (Xiang N. Research on the Actor-Centered Model for the Evolution Mechanism of Buddhist Internet Public Opinion[J]. The World Religious Cultures, 2017(5): 143-149.)

[23] Gulur P, Rodi S W, Washington T A, et al. Computer Face Scale for Measuring Pediatric Pain and Mood[J]. The Journal of Pain: Official Journal of the American Pain Society, 2009, 10(2): 173-179.

[24] Padmala S, Pessoa L. Affective Learning Enhances Visual Detection and Responses in Primary Visual Cortex[J]. The Journal of Neuroscience: The Official Journal of the Society for Neuroscience, 2008, 28(24): 6202-6210.

[25] Nitschke J B, Nelson E E, Rusch B D, et al. Orbitofrontal Cortex Tracks Positive Mood in Mothers Viewing Pictures of Their Newborn Infants[J]. Neurolmage, 2004, 21(2): 583-592.

[26] 刘俊娟, 闫培玲, 肖俊生, 等. 基于复杂网络聚类算法的用户学习行为动态演化模型[J]. 吉林大学学报(理学版), 2025, 63(5):1411-1417.

[27] (Liu J J, Yan P L, Xiao J S, et al. Dynamic Evolution Model of User Learning Behavior Based on Complex Network Clustering Algorithm[J]. Journal of Jilin University (Science Edition), 2025, 63(5):1411-1417.)

基金资助

国家社会科学基金年度项目(25CWW022)

吉林省研究生教育教学改革研究项目(JJKH20230039YJG)

吉林省高教科研重点项目(JGIX24C051)