[Purpose/Significance] To explore the mechanisms of value co-creation in library knowledge services involving multiple stakeholders can promote the innovation and enhance the effectiveness of library knowledge services. [Method/Process] This study analyzed and organized the stakeholders, processes, and influencing factors of value co-creation in library knowledge services. It constructed a theoretical model of value co-creation in library knowledge services and employed the system dynamics method to investigate the mechanisms of the value co-creation stakeholders and their influencing factors. [Result/Conclusion] Simulation results indicate that during the process of value co-creation in library knowledge services, the completeness of the institutional framework has a decisive impact. The greater the number of stakeholders participating in value co-creation, the higher the efficiency of knowledge services and the more significant the improvement in user capabilities. The benefits of value co-creation in library knowledge services become significantly apparent only after a prolonged period of coordination, accumulation, and interaction among multiple stakeholders.

[Purpose/Significance] The term “Tu Shu Guan” in Chinese is the most fundamental and pivotal term in library science in China. Its etymological attributes represent one of the significant issues that cannot be ignored in the study of Chinese library history. The objective of this paper is to elucidate the etymology of the term “Tu Shu Guan” to delineate its dissemination in China and Japan, and to ascertain its linguistic attributes, which will facilitate a more profound understanding of Chinese library history. [Method/Process] This paper used methods of textology, bibliography, and etymology to review the academic history and examine its initial documentary evidence for the term, and examples of its use during the Ming and Qing dynasties. As books were the most important carrier for the dissemination of the term, a systematic analysis of Huang Tingjian’s poems in Japan was conducted to investigate the annotation and publication activities of the poems and to analyze the stereotypes and popularity of the term “Tu Shu Guan” in China and Japan in the modern era. The etymological attributes of the term were then argued from an etymological point of view using the etymological research method. [Result/Conclusion] From an etymological perspective, it can be argued that a term is a “returned word” (Sino-Japanese interactive word), as long as a term originally appeared in Chinese literature, was later introduced to Japan, and was adopted by the Japanese during the Meiji period as a translation of a Western term. Thus, the term “Tu Shu Guan” is not a “Japanese word” but a “returned word,” or “Sino-Japanese interactive word”.

[Purpose/Significance] Accurately and effectively revealing the emotional evolution characteristics and patterns of short video users in sudden natural disaster events is of great significance for related research and practice, such as online public opinion monitoring and short video governance. [Method/Process] First, the LSTM model was used for emotional calculation of short video users. Then, an emotional interaction index for short video users in sudden natural disasters was designed to analyze the degree of emotional interaction among short video users. Finally, time series and visualization methods were comprehensively adopted to reveal the temporal and geographic spatial evolution characteristics and patterns of short video user emotions. [Result/Conclusion] Empirical analysis of 15 sudden natural disaster events in China reveals the spatiotemporal evolution characteristics of short video user emotions. The initial emotions of short video users in different sudden natural disaster events are affected by the severity of the event. The emotional temporal evolution of short video users in different sudden natural disaster events is affected by their daily routines, and a positive correlation exists between short video user emotions and geographic space in different sudden natural disaster events. These findings verify the feasibility and effectiveness of the proposed method in this paper.

[Purpose/Significance] In recent years, emergencies have occurred frequently, which has a seriously impacting social stability and people’s life and property safety. A multi-dimensional analysis of the emergency policy text provides ideas and reference for governments to improve the emergency policy system and formulate effective emergency policies and regulations. [Method/Process] Taking the national policy text as the research object, this artide used the text content and statistical analysis method to analyze these texts from the three dimensions, emergency life cycle-policy, policy tool, policy effectiveness. [Result/Conclusion] The results reveal obvious structural differences in the three dimensions of emergency life cycle, policy tools and policy effectiveness. In terms of emergencies life cycle, it focuses on two stages: prevention and emergency preparedness, emergency response and rescue. In terms of policy tools, booster policy tools account for a relatively high proportion, and carrot-type and whip-type policy tools are rarely used. In terms of policy effectiveness, legal policies are lacking and policy effectiveness is low. In the future, it is necessary to formulate special policy texts for different types of emergencies to strengthen the specialization and refinement of emergency management. At the same time, the application of carrot-type and whip-type policy tools should be appropriately increased to make the emergency response responsibility clear and standardized.

[Purpose/Significance] By conducting text mining on open peer review reports, this study aims to explore the hidden publication bias in the peer review mode and provide theoretical support for its solutions. [Method/Process] The text mining analysis was conducted on the review reports of two modes: Transparent Peer Review (TPR) and Results-Blind Review (RBR). LDA topic analysis, sentiment analysis, and content analysis were used to compare and observe the differences in reviewers’ focal points, emotional tendencies, and the distribution of research results polarity. [Result/Conclusion] Reviewers using the Results-Blind Review (RBR) model tend to focus more on the quality and rigor of research methods and processes, while those using the Transparent Peer Review (TPR) model pay more attention to the results and conclusions of academic papers, and the corresponding evaluations during the review process are more subjective, leading to a potential “results-oriented” publication bias. It draws the following conclusions. Firstly, publication bias in the paper review stage has been confirmed. Second, the core reason behind this bias is the research evaluation system. Third, actively promoting a fully open review process is necessary. Finally, the academic community should shift publishing perspectives and promote emerging publishing models.

[Purpose/Significance] In big data environments, knowledge is ubiquitous, fragmented and large in scale.Recognizing and grasping the existing domain knowledge structures can provide a reference for scientific and efficient domain knowledge organization.[Method/Process] The cognition of domain knowledge structure helped explore the relationships between knowledge units.It showed the principles, connotations, and frameworks of domain knowledge formed by the combinations.Knowledge organized from different perspectives, contexts and applications could present different knowledge structures.The Chinese Library Classification and Chinese Thesaurus has provided hierarchical, equivalent, and related relationships between knowledge units.The former one focused on knowledge category construction to build a tree-like knowledge system, while the latter connected thematic knowledge through concept coordination, related relationships, and multiple memberships, forming a semantic network to express topic related and extended knowledge.The semantic knowledge associative structure constructed the triples to form a semantic network.These networks expressed knowledge related to category affiliation, attributes, and semantic relationships.And they represented implicit knowledge through axiomatic reasoning.[Result/Conclusion] In the big data environment, a tree-like hierarchical structure with multiple affiliations and cross-connections is constructed to represent the category attribution of ubiquitous knowledge.The thematic approach achieves the coarse-grained and conceptual representation of fragmented and fine-grained knowledge, while the semantic relations are more explicit and specific.Semantic associative knowledge structures are more flexible and extensible, which can provide reasoning knowledge.The domain knowledge structures are evolving towards being suitable for knowledge organization in big data environments.

[Purpose/Significance] To conduct an in-depth analysis of the “Strategic Approaches for Libraries Responding to Artificial Intelligence” report published by the International Federation of Library Associations and Institutions (IFLA), this article aims to assist Chinese libraries in achieving service upgrading and innovative development in the era of artificial intelligence.[Method/Process] Using network research and literature analysis methods, this article deeply interpreted the five major parts of the report.Combined with strategic management tools, it conceived a strategic map for libraries to respond to artificial intelligence and consider strategic countermeasures for the library industry in China.[Result/Conclusion] Through in-depth analysis and interpretation of the report, it identifies the three core issues, three limitations, and three characteristics of the report and proposes an innovative strategic map for libraries to respond to artificial intelligence, which includes five levels.Three implications for China’s library industry to respond to artificial intelligence are extracted from the dimensions of policy support, resource positioning, and communication and cooperation.These findings not only deepen the understanding of the report, but also provide valuable theoretical references and practical guidance for the library industry in China to address the challenges of artificial intelligence.

[Purpose/Significance] This paper explores the effectiveness of government rumor-refuting information in the context of significant public health emergencies, focusing on both the process and outcomes.It aims to quantify the government’s ability to manage online rumors, improve the reach and persuasiveness of government-provided information, and offer practical guidance for government public opinion management.[Method/Process] Utilizing a dual perspective of information coverage and persuasion, this paper integrated social network analysis metrics and the theory of information influence to construct a two-dimensional evaluative framework for the effectiveness of government rumor-refuting information.The Analytic Hierarchy Process (AHP) was applied to determine the weights of the indices.The COVID-19 pandemic was taken as a case study for empirical analysis, employing social network analysis, sentiment analysis, and comprehensive evaluation to measure the government’s rumor-refuting efforts across the public opinion cycle of a public health emergency.This approach validated the effectiveness of the evaluative system proposed.[Result/Conclusion] The empirical results demonstrate that the evaluative framework effectively reflects the impact of government rumor-refuting information in the face of major public health emergencies.It recommends that governments develop phased rumor-refuting strategies in response to the progression of public sentiment, innovate in the dissemination pathways of corrective information, and establish a continuous assessment mechanism for the effectiveness of their rumor-refuting efforts.Such actions are crucial for enhancing the influence of rumor-refuting information and elevating the caliber of online rumor management.

[Purpose/Significance] To enhance the semantic accuracy and diversity in data augmentation methods for multimodal rumor detection, exploring models and methods that have the potential to enhance the detection performance can contribute to the identification of online rumors, as well as to the reinforcement of network information governance capabilities.[Method/Process] A multimodal rumor detection model named TARD-GPT-4 was proposed, which leveraged GPT-4 for data augmentation.The model employed BERT and ViT models to extract textual and visual features, respectively.A supervised contrastive learning strategy was used to further explore the label attribute features.Finally, a full connected layer was used for rumor detection discrimination.[Result/Conclusion] Incorporating supervised contrastive learning and prompting large language models using rephrasing method to augment data have a positive effect on improving the accuracy of multimodal rumor detection.Compared to the optimal baseline model, TARD-GPT-4 achieves a 1.62% higher accuracy in multimodal rumor detection.The experimental part also investigates the impact of various data augmentation methods and finds that prompting LLMs for paraphrasing yields the most favorable results.

[Purpose/Significance] This paper examines the geographical spatial characteristics and historical evolution of post station distribution in Gansu Province in Qing Dynasty to provide certain reference for the study of Qing history and digital humanities.[Method/Process] Utilizing ArcGIS and Latent Dirichlet Allocation (LDA), this paper expounded the overall and local spatial distribution characteristics of Gansu post stations in Qianlong, Jiaqing and Guangxu periods in Qing Dynasty, and analyzed the geographical and historical factors influencing these distributions.[Result/Conclusion] The results show that the terrain and geomorphology of Gansu Province play a decisive role in the overall distribution of the stations.And, political, military, economic and social factors jointly affect the geographical spatial distribution characteristics and historical evolution of the stations.

[Purpose/Significance] The rapid development of AI technology has accelerated the research on intelligent information processing of ancient books.In order to grasp the current research and development trend of this field as a whole, it is necessary to systematically sort out the relevant literature so far and provide reference for related research.[Method/Process]Through a systematic review and analysis of existing literature, it defined the connotation and extension of intelligent information processing of ancient books, summarized the main technical methods, tasks and applications, and discussed the development trend.[Result/Conclusion] Intelligent information processing technology for ancient books is based on statistical learning, machine learning, deep learning, pre-training models and large language models.The main tasks include digitization of ancient books, automatic sentence separation and punctuation, automatic word separation, automatic POS tagging, automatic information extraction, automatic classification, citation analysis, machine translation, entity disambiguation, knowledge base construction, and knowledge graph construction, etc.The main research applications focus on specialized and thematic research.In addition, the application of large language model technology, multimodal intelligent information processing tasks and the deepening and expansion of application scenarios are the development trends of intelligent information processing for ancient books.

[Purpose/Significance] Data-enabled intelligent transformation and knowledge discovery have become new growth points for libraries to maintain development vitality and innovate service concepts. The efficient aggregation, orderly organization, integrated utilization and open sharing of massive multi-modal and heterogeneous data in libraries through data governance have become an important driving force for the intelligent transformation of libraries. [Method/Process] Based on the needs of smart libraries in regard to data management, services, and analysis and mining, this article analyzed the connotation, problems and challenges of library data governance under the background of intelligence transformation. [Result/Conclusion] It proposes a project-driven and application-driven implementation of a library data governance system. Based on the smart library management system, the data-enabled library smart transformation goal, and the principles of “data standardization, assetization, knowledge, and intelligence”, it suggests optimizing and controlling data platform architecture, data governance organizational mechanism, data standard specification system, asset management and operation mode, full-process data security and compliance governance, and data integration application. It aims to ensure data governance can better support the development of smart libraries and facilitate the intelligence transformation of library undertakings.

[Purpose/Significance] The fourth-generation evaluation theory, with its core concept of collaborative construction, has driven the change of scientific research evaluation thinking. Under the background of this theory, the evaluation of the research impact also pays more attention to the process of impact generation and dissemination, and emphasizes the interactions of different stakeholders in the process. [Methodology/Process] Driven by both the categorized evaluation system at the policy level and the change of evaluation theory at the theoretical level, this study proposed the categorized evaluation idea of selecting evaluation subjects based on the content attributes of scientific knowledge. It further verified the rationality of this idea through real case data. This study took REF2021 case data as the main source of analysis, and employed prompt learning to assist manual identification of government, enterprise, university and public stakeholders. Based on it, it explored the differences in the participation of different stakeholder types in the research impact generation process in terms of the distribution of participation and cognitive themes. [Result/Conclusion] The results demonstrate differences in the subjective cognition and realistic participation levels of the four stakeholder types in the research impact generation process. On the basis of the confirmed rationality of evaluation ideas, this study proposes suggestions for the selection of evaluation subjects based on disciplinary category from the specific case data, and provides new practical ideas for academic evaluation practices in China.

[Purpose/Significance] This study explores the cognitive transition characteristics of users in the exploratory search process and analyzes the users’ cognitive changes under different task types and search intentions. It contributes to a further understanding of the learning mechanism in the exploratory search process. [Method/Process] An experiment was used to record the participants’ search process on the video screen. The think-aloud method was used to collect the participants’ descriptions of the search process. The questionnaire gathered data on the participants’ personal information, task perception, and task answers before and after the search. Through coding of the video and textual data, the article analyzed the length and probability of users’ cognitive transiting paths across different task types, and the characteristics of these paths with different search intents. [Result/Conclusion] This study finds that the cognitive transition path is the shortest when searchers take procedural tasks. Searchers mostly initiate their cognitive transitions from a state of “understanding”, and application is the demarcation point between ascending and descending cognitive states. Analysis of user cognitive transition paths reveals significant variations across different search intents. Paths associated with information collection and information assessment exhibits the highest degree of complexity.

[Purpose/Significance] This study aims to extract structured item information from free-text clinical scales using ChatGPT without annotations, which efficiently advances the structuring and intellectualization of medical scale resources. [Method/Process] A framework for item information extraction was defined, including eight attribute types and considering the structural differences in clinical scale measurement concepts. A dataset was constructed by collecting 59 commonly used clinical psychometric assessment scale documents. Zero-shot prompt templates were designed based on measurement concept levels, and experiments were conducted using the official ChatGPT-3.5 and ChatGPT-4 interfaces. The extraction performance and possible influencing factors of different ChatGPT versions in processing different clinical scale texts were analyzed from multiple perspectives. [Result/Conclusion] The extraction performance for scale item sources is the best, with Micro-F₁ and Macro-F₁ scores of at least 98.90% and 97.83%, respectively. This is followed by response options, instructional guidance, and scoring rules, with item numbers and instructions showing moderate performance. Clinical explanations have the lowest performance, with Micro-F₁ and Macro-F₁ scores of 47.73% and 45.51%, respectively. ChatGPT-4 performs better overall, but the recall rate of some attributes is weaker than that of ChatGPT-3.5. The increase in measurement concept levels, dimensionality, number of items, and text length are found to reduce model performance. In summary, ChatGPT can efficiently assist in the structuring of medical scale resources, especially when dealing with simple scales.

[Purpose/Significance] Academic journals are widely used as a processing unit for identifying discipline structure. By identifying discipline subfields through journal clustering, it is possible to effectively reveal the intrinsic structural features of a discipline. From journal contents on the semantic correlation of articles, it proposes a new method of constructing correlation between journals, called journal article topic coupling, and a discipline structure identification model. [Method/Process] It employed the LDA topic model to identify the topic and its topic distribution. Based on this, it proposed a threshold-based method to construct journal article topic distribution, and adopted the t-SNE algorithm to obtain low-dimensional embedding of the high-dimensional distribution to address the difficulty of clustering journal by high-dimensional journal topic distribution. [Result/Conclusion] It conducts empirical analysis on 68,078 articles in 107 economics and management journals included in CSSCI from 2018 to 2022, and describes the model’s parameter selection and setting methods. The results demonstrate that the new method effectively utilizes article topic correlation on the semantic dimension for journal clustering and discipline subfields identification, divides domestic economic and management disciplines into 16 subfields, including 9 in economics, 5 in management, and 2 in the cross-field, along with an interpretation of the research scope of each subfield.

[Purpose/Significance] The current subscription fees for literature information resources in China are offered at discounted rates. But in the continuous development of global open access, China needs to pay high conversion fees to complete the comprehensive transformation of open access. University libraries are facing new challenges in the construction of literature information resources. [Method/Process] This paper conducted research from the perspectives of open access influence comparison, the transformative policies of publishing groups, and the funding of university literature resources. It proposed countermeasures research, including the construction and management of university publication information, calculation of open access transformation, and construction of open access transformative mode. [Result/Conclusion] University libraries should be cautious in open access transformation. The subscription-based publishing model has generally achieved a balance, and teachers and students can publish for free in hybrid journals, therefore, this paper proposes prioritizing support for publishing in high-quality, fully open access journals. Each university needs to carry out the management of its own publication costs, and conduct a comprehensive evaluation based on specific publisher policy, the number of university publications, and the price system of article processing charges. It provides direction guidance and case reference for university libraries to respond to the open access transformation.

[Purpose/Significance] This study aims to explore the specific mechanisms and implementation methods of value co-creation in university library user services, providing new perspectives for their sustainable development. [Method/Process] Based on the review of the concept and development of value co-creation, this study selected the comprehensive service center in Peking University Library as a case study. It focused on four aspects: the innovation of user-supported service concepts, the planning of institutional frameworks for resource integration, the construction of open and integrated service environment, and the development of a collaborative service ecosystem among participants. [Result/Conclusion] Actively promoting value co-creation in university library user services can better meet the users’ diverse needs and promote the sustainable development of library services.

[Purpose/Significance] This article investigates the cognitive and emotional factors influencing users’ willingness and behavior to disclose false information on social networks, so as to promote the healthy development of social networks. [Method/Process] It explored the complex mechanism of users’ disclosure willingness and behavior from the perspective of dual paths of cognition and emotion. It analyzed the data from 847 valid questionnaires using SEM and fsQCA. [Result/Conclusion] ①Engagement, perceived value, perceived risk, and self-efficacy positively affect the disclosure willingness and the disclosure behavior.②Emotional arousal positively influences users’ willingness to disclose information. Negative emotion leads to a stronger willingness to disclose false information. ③Cognitive factors make users more focus on information quality, while emotional factors make users more focus on the reputation of information sources. ④Self-efficacy and emotional arousal show positive regulation in the influence of disclosure willingness on disclosure behavior. ⑤Two paths triggering users’ willingness to disclose false information are found, with self-efficacy as the core factor.

[Purpose/Significance] It is crucial to explore the factors and pathways that render social media users susceptible to health misinformation. This helps understand the underlying psychological mechanism of these susceptible users when facing health misinformation, thereby aiding in enhancing their ability to identify such misinformation. [Method/Process] This study collected data through an online situational experiment. Based on the MOA model and the ELM theory, it employed fuzzy-set Qualitative Comparative Analysis (fsQCA) to investigate the configuration paths influencing social media users’ susceptibility to health misinformation. [Result/Conclusion] The results indicate that three configural types of users who are susceptible to health misinformation: the health-consciousness core-driven type, the popularity- driven core type, and the dual-driven type characterized by both high health consciousness and information popularity. Among these, high health-consciousness and the reliance on information popularity-based pathways emerge as pivotal factors influencing health misinformation susceptibility.

[Purpose/Significance] While the “renaming” of the “Library, Information, and Archives Management” has come to an end, how to reconstruct its sub-disciplines under the first-level discipline framework of “Information Resource Management (IRM)” is one of the important topics in front of the domestic academic community. The emergence of the “Information Resource Management Science (IRMS)” concept provides a new direction for the research on this topic. [Method/Process] The article adopts theoretical analysis method to explore the system structure of IRMS in depth from four aspects: discipline system, theoretical foundation, method system, and sub-discipline system. [Result/Conclusion] The research sorts out the discipline system of IRMS, improves its theoretical foundation, and method system. Based on the conceptual origin of IRMS as reflected in the two book series, the Modern Information Resource Management Series and Information Resource Management Series in the Age of Digital Intelligence, as well as the inheritance and innovation of “Library, Information, and Archives Management”, which is the main part of IRMS, it constructs two distinct sub-discipline systems.

[Purpose/Significance] This study aims to construct a comprehensive assessment framework to objectively measure, compare and analyze the quality and dynamic evolution trends of government open data in various regions. This framework will then be used to identify potential problems and solutions related to government open data quality management, providing reference for government data service optimization and data governance decision-making. [Method/Process] Relying on multi-indicator comprehensive evaluation theory and Delphi method, this article formulated government open data quality assessment indicators and their dimensions. Using the single-sample Wilcoxon test and combined weighting method, the availability, weights, and dimensional scores of the assessment indicators were analyzed to construct a framework for assessing the quality of government open data. [Result/Conclusion] This article evaluates the quality of open government data on the government websites in the sample areas in five dimensions: content, expression, access, protection and quality of use. Based on it, it analyzes existing problems, and proposes optimization countermeasures.

[Purpose/Significance] In the trend of sensory regression in digital age, libraries need to pay attention to sensory participation and sensory design in services to enhance the value and service effectiveness. [Method/Process] Using explanatory multi case study methods, it constructed an analytical framework based on sensory design and service design theories to explain the STEAM education cases in libraries and to verify and improve the theoretical framework. [Result/Conclusion] Through case studies, the study improves the theoretical framework of sensory design for library services. Starting from the dimensions of resources, users, and works, it analyzes application scenarios of sensory design in library service design and offers suggestions.

[Purpose/Significance] Large language models trustworthiness can ensure their reliable and safe deployment in key fields such as healthcare, education, and law, and gain the public trust. It is important to promote the healthy and sustainable development of generative artificial intelligence technology. [Method/Process] This paper uses the literature research to analyze the relevant literature on the large language models trustworthiness. It comprehensively analyzes the large language models trustworthiness from four aspects: its concept and connotation, its influencing factors, the challenges, and the “TRUE” strategy. [Result/Conclusion] The research on the large language models trustwortiness is still in the exploratory stage. Starting from its connotation, influence, challenges, it establishes a “TRUE” strategy system of information technology, information resources, information users and information environment, in order to promote the reasonable use of LLM and drive the development of generative artificial intelligence towards a better future.

[Purpose/Significance] In the current context of frequent major emergencies, the pivotal role of social media has become evident. Although its intelligent recommendation algorithms have increased the visibility of disaster information, they might also exacerbate the spread of internet rumors and the fermentation of group information herd behavior. Understanding the influencing factors of group information behavior in major emergencies under the intelligent recommendation algorithms can reveal the tendencies and decision-making mechanisms of group information following, enabling the formulation of more effective crisis management and emergency plans. [Method/Process] This paper combined content coding analysis with fuzzy-set Qualitative Comparative Analysis (fsQCA) methods, adopting an exploratory-then-confirmatory followed by a verification research strategy, and designed a sequential mixed research method. It identified the key factors influencing group information herd behavior in the context of major emergencies based on social learning theory. The fsQCA method was used to construct a configurational path analysis framework. [Result/Conclusion] The study identifies eight key factors influencing group information herd behavior during major emergencies, including observational learning, perceived social influence, recommendation algorithmic affordance, recommendation algorithm experience, perceived social value, perceived social identity, self-regulation, and self-efficacy. Based on these factors, it discovers three distinct types of information herd behavior in two categories: information herd behavior driven by herd mechanism factors, information herd behavior involved by triadic interactions and imitation of others, information herd behavior involved by triadic interactions and discount own information, and non-herd information behavior. These findings offer new perspectives for a deeper understanding of the role of social media in responding to major emergencies and provide theoretical support for optimizing intelligent recommendation algorithms on social media platforms in specific scenarios.

[Purpose/Significance] The analysis of public speech acts and sentiment dissemination in natural disaster crisis helps understand the acts intentions, acts patterns and psychological states of government social media users, and provides theoretical support and reference basis for governmental crisis management and public opinion control in crisis events. [Method/Process] Taking the “Typhoon Dusuri” event in 2023 as an example, this research integrated qualitative content coding, lag sequential analysis, cluster analysis, theme mining and other research methods to meticulously analyze 254 blog posts and 5036 comments from five government microblog accounts. [Result/Conclusion] In the natural disaster crisis, public speech acts are mainly concentrated in expressive and declarative categories, and there are three main speech act modes, information, relationship and sentiment at the same time. The public sentiment mainly shows homogeneous infection and negative extended transmission, characterized by temporal stages. Public sentiment levels vary across speech act modes, with information mode in lower sentiment level. The “action and rescue” blog topics significantly increase the public sentiment in all the three modes. Due to the dynamic nature of public participation in governmental social media and the difficulty of portraying the participation process in detail, the speech acts classification, lag sequence analysis and automatic clustering methods used in this study provide new perspectives and methods for public participation in crisis governance research.

[Purpose/Significance] Exploring the influencing factors and mechanisms of knowledge service effectiveness of science and technology think tanks can comprehensively enhance their service effectiveness, promote sustainable development, and maximize their value in decision-making consultation. The research will provide a reference for designing the scale and evaluation system of the knowledge service effectiveness of science and technology think tanks. [Method/Process] In this paper, it investigated the current status of knowledge services in science and technology think tanks through in-depth interviews on programmatic grounded theory. The collected raw data from the interviews were analyzed by three-level coding. Finally, it constructed the theoretical model of influencing factors and mechanisms of knowledge services effectiveness in science and technology think tanks. [Result/Conclusion] It finds that the knowledge service effectiveness of science and technology think tanks includes three aspects: the realization of application value, the deepening of theoretical research, and the formation of supply and demand mechanism. Basic resources (supporting force), wisdom synergy (security force), user demand (driving force), and service characteristics (constructing force) jointly determine the level of their knowledge service effectiveness. There is a significant mutual promotion and dependence relationship among the dimensional factors. Among them, user demand exhibits a radial characteristic in its influence on the effectiveness of knowledge services of think tanks and other factors, tending to the center of knowledge services. The demand-driven synergistic integration of diversified institutional resources has a significant impact on the think tank services effectiveness. It is conducive to easing the tense supply and demand relationship between “think tank-decision-making”. This idea need to be emphasized in future relevant policies and service practices.

[Purpose/Significance] Exploring the innovative development of red culture reading promotion in university libraries from the perspective of converged media is of great significance for relevant institutions to fully leverage the advantages of converged media communication matrix and carry out practical activities for promoting red culture reading. [Method/Process] A comprehensive survey was conducted on the current situation of promoting red culture reading in 42 “Double First Class” university libraries through online research, face-to-face interviews, literature reviews, and telephone interviews. The survey results were classified, statistically analyzed, and a framework was constructed by drawing on stakeholder theory and the logic of social marketing theory. [Result/Conclusion] This study identifies six key factors influencing the red culture reading promotion in university libraries including red resources, space, activities, stakeholders, communication media, and performance evaluation. It proposes an innovative path for the integration of converged media and red culture reading promotion from the aspects of red cultural resources integration, spatial integration, multi-activity integration, publicity integration, and interest integration.

[Purpose/Significance] This study aims to solve the problems of inconsistent standards and incomplete evaluation indicators of scientific and technological documents in the current Traditional Chinese medicine field in China. This article has established a clinical application value evaluation system, and has designed an evaluation system for scientific and technological documents in the field of Chinese medicine to help researchers screen high-quality literature and promote the progress of clinical research. [Method/Process] Based on the standardization of literature, the characteristics of measurement, the characteristics of traditional Chinese medicine, and the clinical practice guide, the evaluation indicator system was established. The Delphi method and Analytic Hierarchy Process (AHP) were employed to optimize the indicators and determine their respective weights. Indicators were identified by combining rule-based methods and deep learning. The value score of the clinical literature was obtained through weighted summation. Finally, 10 relevant articles were selected to test the effectiveness and feasibility of the system. [Result/Conclusion] The experimental results show that the scores generated by the system can reflect the clinical application value of the literature, the evaluation system is scientific and effective.