Preliminary Conference Agenda

突发公共卫生事件成为了21世纪威胁人类健康安全的重大健康危机之一。在抗疫时期，网络成为了突发公共卫生事件时期群众了解信息、与人沟通的主要渠道。当不同的观点、想法、情绪在互联网中相互碰撞时就极易引起网络舆情风险。因此，在全球性的突发公共卫生事件频发的背景下，做好网络舆情风险的管理与控制具有重要意义。本文提出了一个突发公共卫生事件网络舆情风险分级方法。在指标体系的构建部分，利用等级全息建模（Hierarchical Holographic Modeling，HHM）框架识别突发公共卫生事件网络舆情风险因素并构建指标体系；在多准则决策方法部分结合层次分析法排序II（Analytical Hierarchy Process Sort II , AHPSort II）和摇摆赋权（Swing Weighting , SW）方法以及模糊理论，提出了一种新的模糊多准则决方法。该方法被应用于突发公共卫生事件时期微博平台的网络舆情风险研究。最后，根据研究结果，提出了一些有效的舆情风险管理建议。本文可以为政府在今后可能发生的突发公共卫生事件中制定或完善相关舆情风险治理政策提供参考。

为实现茶文化信息资源在语义层面的整合，识别其文化基因对其知识内涵进行细粒度挖掘和逻辑关联建构对于茶文化的现代化传承和发展具有重要意义。本文尝试将文化基因理论融入茶文化信息资源的知识重构中，通过将文化信息资源中的茶文化基因进行提取、识别，引入知识元组织方法进行知识关联形成知识资源，依靠知识图谱从多维度完成对茶文化信息资源知识服务实现路径的探讨。通过实证研究，验证了文化基因理论融入文化信息资源进行语义关联分析的适用性与可行性，知识资源的细粒度解构和重组方案可以为茶文化领域的多维重组与融合应用提供借鉴，也为中华优秀传统文化的系统性保护和文化基因的赓续发展提供参考。

科研合作预测一直是学界广泛关注的问题。科研合作网络同时具备社会网络和知识网络双重特征，使得准确预测合作关系成为挑战。现有研究大多通过探究文本内容和网络结构两大特征，利用链路预测方法预测潜在合作关系。早期大量研究关注基于相似性度量的合作预测，随着网络表示学习方法的兴起，学者开始关注基于表示学习的链路预测，利用文本表示学习和网络表示学习方法捕捉文本内容特征和网络结构特征。然而，如何有效融合科研网络结构和文本内容特征以提高科研合作预测性能仍是一个挑战。基于以上问题，本文提出了一种基于表示学习和图注意力网络的科研合作预测模型。具体来说模型包含以下四个阶段：第一，通过有偏的随机采样策略，构建基于学者合著关系的科研合作网络；第二，通过DeepWalk、BERT等网络表示学习和文本表示学习方法，抽取学者研究内容特征和网络结构特征；第三，采用注意力机制获取注意力系数，并将研究内容特征和网络结构特征输入到设计好的图注意力网络GAT中，利用邻居聚合策略融合两类特征以获得更新后的学者综合表示；第四，使用学者综合表示计算节点间的相似性，并基于随机森林进行科研合作预测。最后，本文通过多种基线实验和消融实验，在多个子数据集中证明了方法的有效性。

As an important center for the protection and research of global cultural heritage, the British Library holds a rich collection of global manuscripts and has undertaken numerous outstanding manuscript digitization and utilization projects. Taking the "Hebrew Manuscripts Digitization Project" as an example, this paper employs a case study methodology to interpret the distinctive features of manuscript digitization within the project, focusing on the development of a standardized manuscript digitization process and the use of a convenient metadata management application. The paper further uncovers the project's experiences in utilizing digitized manuscripts in four aspects: constructing a digital manuscript platform, developing 3D manuscript models, setting up virtual manuscript exhibitions, and establishing digital manuscript awards. Based on these findings, this paper recommends that Chinese ancient book preservation institutions should start from establishing specialized funding organizations, seeking international cooperation, standardizing digitalization standards for ancient books, innovating in display forms of ancient books, and setting public participation awards to digitize and utilize precious Chinese ancient books, thereby promoting the inheritance and enhancement of Chinese culture.

摘 要：[目的/意义]基于斯宾诺莎情感理论进行弹幕视频的方面级情感演化分析可以为视频制作者评价科普效果、提升视频内容质量提供自检方法，也能为舆情督导部门鉴别非理性弹幕提供一定参考。[方法/过程]以斯宾诺莎情感理论为基础，基于弹幕与视频内容的对应关系构建时间序列模型，以B站人工智能科普类视频为例，运用LDA主题聚类、情感分类、可视化等多种方法，讨论了各个主题的弹幕匹配程度并对每个主题进行方面级情感分析，根据该结果分别绘制了一般情感趋势曲线和主动情感趋势曲线，并进行对比分析。[结果/结论]研究结果表明，在方面级情感分析过程中，观众的弹幕内容从被动的消极情感逐渐走向主动的积极情感；引入斯宾诺莎情感理论的情感演化分析得到的演化趋势相较于传统的演化趋势更加清晰，且能够展示出传统分析中被忽略的影响因素。

Exploring the archives’ image in the eyes of Generation Z holds significant value for archives in shaping their future image management and fine-tuning public services. Employing the Destination Image Theory, this research utilized data from the Xiaohongshu platform to unveil how Generation Z perceives archives from both cognitive and emotional perspectives. In June 2023, researchers collected 1000 notes related to "archives" using Python, followed by content analysis utilizing Weiciyun platform and MAXQDA. The findings indicate that in terms of cognitive image, archives, as public spaces, encompass both functional and leisure aspects, with Generation Z showing particular interest in transportation location, architectural design, infrastructure, and archival resources as spatial components. In the realm of emotional image, Generation Z, while categorizing their emotional tendencies towards various facets of archives into "positive emotion-oriented" "neutral positive emotion-oriented" and "neutral emotion-oriented" overall demonstrates a positive emotional inclination. For Generation Z, archives is a niche but unexpected treasure space.

With the acceleration of the digitization process, Internet misbehavior has brought various negative effects on citizens' property, physical and mental health, and has become a major factor that harms social order and hinders national security. The practice of countries around the world shows that improving citizens' digital literacy is an effective way to solve these problems, and constantly enhancing the digital literacy and skills of the whole people and bridging the digital gap are also the realistic needs of the current situation in China. On the basis of social constructivism, this study explores the connotation framework of "citizen digital literacy" suitable for the Chinese context, constructs the internal and external dual-cycle structure of citizen digital literacy, and extracts 12 components of citizen digital literacy framework in the Chinese context by combing the relevant digital literacy framework and analyzing the needs of five contexts, including politics, economy and culture. It is explained in detail from three dimensions of knowledge, skill and attitude. Finally, based on the perspective of government attention, this paper analyzes the content of 139 digital literacy related policy texts, judges the current policy focus, and puts forward implementation suggestions from two aspects: balancing the policy focus and adjusting the policy focus.

文章立足于中华文化走出去的时代使命，探索中国图书馆学教育文化影响的叙事方法。采用文献研究法和历史研究法，在阐述国内外对文化影响的代表性理论研究成果基础上，结合教育与文化的关系和教育对文化影响的理论成果，构建起中国图书馆学教育文化影响的叙事框架。中国图书馆学教育对文化影响叙事方法主要包括图书馆学教育对文化的保存传承、选择传播、融合发展和创新创造4个维度。

[Purpose/Significance] The open government data platform is a carrier for online access and acquisition of government data, and how to improve the utilization level of the platform is an important issue facing the promotion of open government data. [Method/Process] Taking 16 provincial open government data platforms in China as the research object, combining the TOE framework, introducing the necessary condition analysis method, combining NCA and fsQCA, analyzing the two types of causality, necessary and sufficient, and exploring the key factors and effective paths that affect the utilization level of China's provincial open government data platforms. [Result/Conclusion] The study finds that there is no single necessary non-sufficient condition that affects the utilization level of provincial open government data platform, and that three antecedent variables, namely, data analysis technology capability, data construction level, and organizational readiness, are the key factors affecting the utilization level of open government data platform. Four conditional grouping patterns that generate high utilization levels of provincial open government data platforms are found using the grouping perspective and summarized as data technology and platform construction driven under government-led logic, technology and environment driven under demand response relying on financial resources, organization and environment driven under financial assistance relying on platform construction, and environment driven under government-led logic relying on data technology.

Exploring the knowledge flow of micro-level knowledge entities between different disciplines is conducive to reflecting knowledge contributions more comprehensively and objectively. In this paper, a discipline dominance index is proposed to identify the discipline ownership of knowledge entities at the micro level, analyze it from a global perspective, and then quantitatively measure the interdisciplinary knowledge flow. The results show that biology, clinical medicine, physics, computer science, and engineering occupy the core of the network and export more knowledge to other disciplines. In the fields of social science and humanities and arts, the knowledge contributions of psychology, sociology, and management are relatively high. The idea of measuring knowledge flow at the micro level proposed in this paper can enrich the measurement dimension of knowledge flow.

[目的/意义]旨在研究影响用户采纳冲突性健康信息的因素，有助于了解复杂信息环境下用户的信息决策、信息采纳过程，减少冲突性健康信息对用户认知的负面影响，帮助作出科学的采纳决策。[方法/过程]首先，通过文献扎根与编码析出影响用户冲突性健康信息采纳行为的因素；其次，运用DEMATEL方法分析影响因素与关键程度，从中心度与原因度两方面揭示关键影响因素；然后，通过ISM方法构建影响因素多级递阶结构模型，确定各层级因素之间的影响关系；最后，从用户、健康信息与社会支持等方面提出帮助用户科学应对冲突性健康信息的建议。[结果/结论]影响用户采纳冲突性健康信息的关键原因因素包括用户基本特征、经验积累与健康信息类型；关键结果因素包括用户心理情感、感知权衡与健康信源可信度；重要性因素包括用户心理情感、感知利弊、情感支持与健康信源可信度、信息支持。冲突性健康信息采纳行为是表层因素、中层因素与深层因素共同作用的结果。可以通过提升用户的信息辨别能力、改善健康信息质量、增强社会支持等方面帮助用户科学采纳冲突性健康信息。