Special Topics

257 Parsing and Layout Insights of Agentic Document and Information Service Technology Based on IARPA's Project Announcements Hot! Fu Yun, Liu Xiwen DOI: 10.3772/j.issn.1000-0135.2025.03.001 Clarifying the connotation and scope of Agentic Document and Information Service Technology (ADIST) is crucial for the strategic planning of Document and Information Service (DIS) institutes and the research direction of scholars. This study explores a fine-grained technical content analysis of 87 project announcements published by IARPA to systematically reveal the structure and characteristics of ADIST across four hierarchical dimensions: topic, subtopic, technical issue, and evaluation metric. Accordingly, this study identifies what ADIST is and its key components. To achieve this, a technology layout analysis framework based on project announcements was proposed. A Project Description Model (PDM) was designed incorporating three categories of knowledge elements: research objectives, technical questions, and evaluation metrics. A project announcement text-parsing prompt tailored for PDM was developed, achieving an average recognition accuracy of 92.94% for the three knowledge elements when applied in GPT-4o (Generative Pretrained Transformer 4 Omni). Additionally, TopicGPT was used to construct hierarchical topics (topic-subtopic) based on the summary and research objective texts of project announcements. This study further reveals the technical layout content and characteristics by integrating hierarchical topics, technical issues, and evaluation metrics. The analysis concludes that ADIST refers to intelligent technologies applied to DIS tasks and information technologies that enhance the agentic-driven transformation of the DIS workflow. It encompasses four key aspects—intelligent data, intelligent computing, intelligent cognition, and intelligent systems—with four core characteristics and four evaluation principles. Case studies further validate the reliability of these findings. Finally, guided by a broad DIS perspective, this study proposes six key future-oriented tasks for ADIST development. These include three key technical research challenges: intelligent analysis and cognitive modeling for DIS application scenarios, goal-driven multimodal DIS data production and organization, and intelligent information technologies for complex DIS scenario computation and analysis. Additionally, three core practical application challenges are identified: standardization and automation of DIS workflows, usability and dissemination of DIS tools and analytical results, and determinacy and measurability of DIS evaluation frameworks. 2025 Vol. 44 (3): 257-270 [Abstract] ( 1 ) HTML (1 KB)  PDF (2474 KB)  ( 2 )

Intelligence Theories and Methods

271 Smart Government Information Decision-Making for Major Emergencies: Cognitive Framework and Practical Path Hot! Pang Yufei, Zhang Haitao, Li Yilin, Zhang Chuanyang DOI: 10.3772/j.issn.1000-0135.2025.03.002 From the perspective of information resource management, with a focus on the precise scenario of major emergencies, this study investigated the cognitive framework and practical path of smart government information decision-making. First, based on a literature review and system theory thinking, the connotations, tasks, characteristics, and practical logic of smart government information decision-making were analyzed and defined, and a comprehensive cognitive framework for smart government information decision-making was constructed from a panoramic perspective. Second, the practical path of smart government information decision-making is summarized from four aspects: multi-source and multimodal information fusion acquisition, digital intelligent empowerment of information processing, execution of information decision-making plans, and evaluation of information decision-making quality and efficiency. This study integrates information wisdom and missions into the implementation process of smart government decision-making practices, opens new perspectives on optimizing and enhancing the decision-making structure and capabilities of smart governments, helps achieve the goal of precise decision-making in smart government science, and provides a certain reference for the modernization transformation and upgrading of government governance capabilities and systems in the new era. 2025 Vol. 44 (3): 271-281 [Abstract] ( 2 ) HTML (1 KB)  PDF (1845 KB)  ( 2 )

282 Knowledge-Discovery Method Driven by the Collaboration of Data and Knowledge: Concept, Mechanism, and Model Hot! Yao Sumei, Lu Quan DOI: 10.3772/j.issn.1000-0135.2025.03.003 Knowledge discovery is a critical theoretical framework for addressing the challenges posed by vast amounts of data and complex problems, advancing scientific research and enhancing decision support capabilities. “Data” and “knowledge” are core concepts in information science, and knowledge discovery driven by data or knowledge serves as an essential approach to solving research problems in data-intensive or knowledge-intensive contexts. However, pervasive issues of imperfect data and uncertain knowledge limit the effectiveness of these methods. The co-driven approach offers an innovative pathway for discovering new knowledge through the complementary integration of data and knowledge. Despite its potential, a comprehensive and in-depth analysis of co-driven methods remains insufficient. This study adopts a cognitive logic structure of “what,” “why,” and “how” to explore the basic concepts, mechanisms, and models of knowledge discovery driven by the collaboration of data and knowledge. First, it introduces the fundamental concept of knowledge discovery through data-knowledge co-driven mechanisms, along with a detailed explanation of the newly introduced concepts of imperfect data and uncertain knowledge, which are essential components of this framework. Subsequently, the mechanism section examines the multi-path and multi-objective strategies for integrating data into knowledge-driven knowledge discovery and integrating knowledge into data-driven knowledge discovery. It explains the essence and operational mechanisms of co-driven knowledge discovery by emphasizing the cross-complementarity between data and knowledge. Finally, this study proposes a problem- and scenario-driven basic model of knowledge discovery that is co-driven by data and knowledge. It elaborates on three primary categories of internal modeling for co-driven knowledge discovery: predominantly knowledge-driven discovery (construction and error-correction modes), predominantly data-driven discovery (embedding, correction, and guidance modes), and other collaborative knowledge-discovery methods (hybrid and concurrent modes). The co-driven knowledge-discovery approach, which encompasses multiple co-driven modes, balances the complementary and synergistic effects of data and knowledge, thus providing a more comprehensive framework and process for knowledge discovery. This approach expands methodological innovation and problem-solving perspectives within the discipline of information resource management. 2025 Vol. 44 (3): 282-295 [Abstract] ( 2 ) HTML (1 KB)  PDF (2249 KB)  ( 1 )

296 Intentions of Data Papers in Open Science: A Case Study in the Biomedical Field Hot! Wang Lijun, Liu Ru, Yang Bo, Liu Zhihui, Zheng Ming DOI: 10.3772/j.issn.1000-0135.2025.03.004 This study investigates the citation characteristics and trends of data papers to explore their practical utility, using a full-text corpus of citations. Focusing on the data-intensive biomedical field, we analyze citation intentions across three dimensions: time, space, and journal distribution. Utilizing automatic identification methods, the findings reveal that the citation time of data papers is highly consistent with their publication time. Moreover, the proportion of substantive citations is slightly higher than that of non-substantive citations. In substantive references, citation sentences are primarily concentrated in the data/method section. The number of papers describing datasets is more than those describing databases; however, the latter incorporates higher and more sustained citations. 2025 Vol. 44 (3): 296-308 [Abstract] ( 2 ) HTML (1 KB)  PDF (3368 KB)  ( 1 )

309 Quantitative Evaluation and Optimization of GAI Policy and Regulation Texts' Focus on Information Governance Hot! Wang Xu, Liu Binbin, Qiu Junping DOI: 10.3772/j.issn.1000-0135.2025.03.005 In the digital and intelligent era, generative artificial intelligence (GAI) has been garnering worldwide attention. The emergence of large language models has triggered many chaotic phenomena in the information ecology. The quantitative evaluation and optimization of GAI policies and regulations have practical significance in promoting the study of GAI governance rationalization, facilitating the enhancement of the level of social risk management and the effectiveness of information governance, and advancing national cyberspace governance. First, this study analyzes the information governance dilemma triggered by GAI content. Second, it employs the policy modeling consistency (PMC) index model method and combines it with the MatLab tool to quantitatively evaluate and analyze the texts of 14 global GAI policies and regulations. The findings reveal that the overall level of consistency of policies and regulations is relatively good, but there are still problems with unclear types of services in legislative industries and areas, restricted application functions of trustworthiness and controllability, and solidification of the scope of technical safeguard governance. Accordingly, this study proposes four optimization dimensions—technology optimization, risk assessment, application deployment, and international policy and regulation integration. Owing to the information governance dilemma triggered by GAI, the study refines agile governance into three core dimensions—intelligent services, trusted applications, and technical security—which serve as a yardstick for evaluating GAI policies and regulations. It constructs an optimization framework of flexible solutions and scenario-based hierarchical governance modes with both soft and hard methods, and proposes an optimization proposal of GAI policies and regulations oriented to information governance. 2025 Vol. 44 (3): 309-324 [Abstract] ( 4 ) HTML (1 KB)  PDF (4068 KB)  ( 1 )

Intelligence Technology and Application

325 Can Combinative and Disciplinary Novelty Enhance the Technological Impact of Scientific Papers? A Dual Perspective of Direct and Indirect Technological Impact Hot! Yang Alex J., Liu Meijun, Bu Yi, Zhao Star X., Deng Sanhong DOI: 10.3772/j.issn.1000-0135.2025.03.006 Understanding whether combined novelty and interdisciplinary knowledge integration can enhance the technological impact of scientific papers is important for the development of new quality productive forces. To this end, this study employs Monte Carlo simulations to measure combinative novelty using higher-order metrics derived from real and simulated distributions of knowledge combinations. Disciplinary novelty is quantified using co-citation matrices and distributional distances across disciplines. Adopting a dual perspective of direct and indirect technological impact, this study integrates a “paper-to-patent” citation network with a deep “paper-to-paper” citation network to propose metrics such as patents directly and indirectly citing papers. These metrics capture the direct and indirect technological impact of scientific papers. Based on a fixed-effects regression analysis of 30 million scientific papers from the Microsoft Academic Database, the findings reveal that combinative novelty positively promotes both direct and indirect technological impacts of papers, with a stronger effect observed on indirect technological impact. By contrast, disciplinary novelty exclusively promotes indirect technological impact. Furthermore, the study identifies that, in STEM fields and large collaborative teams, both combinative and disciplinary novelty effectively enhance the dual technological impact of papers. Notably, the positive effect of disciplinary novelty on the indirect technological impact of papers exhibits a declining trend over time. 2025 Vol. 44 (3): 325-338 [Abstract] ( 2 ) HTML (1 KB)  PDF (3128 KB)  ( 1 )

339 Measuring the Novelty of Scientific Papers Using Cross-dimensional Feature Integration Hot! Ma Ming, Zheng Zhejun, Mao Jin, Bai Yun, Li Gang DOI: 10.3772/j.issn.1000-0135.2025.03.007 Accurately assessing the intrinsic novelty of scientific papers is essential for advancing academic research and ensuring high-quality evaluation of scientific innovations. This study introduces a method for measuring the novelty of scientific papers that integrates cross-dimensional features based on their knowledge content and academic communication structure, enhancing the precision of academic assessments. First, a structured knowledge representation model for scientific papers is constructed using a specific combination of questions and methods, with a domain-pretrained language model employed to weigh these combinations. Second, from the perspectives of knowledge content and academic communication structure, we construct a cross-dimensional comprehensive measurement index to evaluate the novelty of scientific papers, focusing on ex ante features such as originality, complexity, and research popularity. The effectiveness of the proposed method is validated through empirical analysis of a biomedical dataset and ex post impact verification of novel papers. The empirical analysis results demonstrate that the proposed method is resilient to time and environmental factors, maintaining its effectiveness over long-term spans and successfully uncovering novelty patterns of papers in a specific field. Furthermore, comparisons with single-dimensional methods show that the proposed method synthesizes and captures multidimensional composite features more effectively, avoiding the oversimplification and one-sidedness of measurement dimensions. This study introduces new perspectives and methodologies for measuring the novelty of scientific papers and offers researchers a valuable tool for identifying and advancing innovative research. 2025 Vol. 44 (3): 339-352 [Abstract] ( 3 ) HTML (1 KB)  PDF (2553 KB)  ( 1 )

353 Research on Disruptive Technology Topic Recognition Based on CLIP-LDAGV Multimodal Information Fusion — A Case Study of the New Energy Field Hot! Lyu Kun, Zhang Weixu, Jing Jipeng DOI: 10.3772/j.issn.1000-0135.2025.03.008 Currently, global technological innovation is exhibiting a trend of rapid development and high integration. Accurately identifying disruptive technological themes that drive comprehensive innovation has become a key driving force for scientific and technological development and economic growth. However, traditional methods of subversive technology topic identification rely primarily on single-modal data, which have some limitations. This study constructs a news text and image feature fusion vector based on the CLIP (contrastive language-image pre-training) and LDAGV (linear discriminant analysis & global vectors for word representation) models, and uses k-means clustering iterations combined with three disruptive technology topic indicators for screening, achieving the fusion of multimodal information and accurate identification of topics. Using the new energy field as an example, the feasibility and effectiveness of the model for disruptive technology topic recognition are verified. Compared with other single-modal models, multimodal information fusion models have additional advantages in identifying disruptive technology topics. 2025 Vol. 44 (3): 353-368 [Abstract] ( 3 ) HTML (1 KB)  PDF (5095 KB)  ( 1 )

Intelligence Discipline Development and Construction

369 Information Science and Information Service Enabling New Quality Productive Forces: A Review of the 2024 Annual Conference on Information Science in China Hot! Li Yuelin, Pan Zhengyuan, Fan Sinuo, Zhang Xiangyihong DOI: 10.3772/j.issn.1000-0135.2025.03.009 The 2024 Annual Conference on Information Science in China, Information Science and Information Service Development Forum, and 14th National Information Science Doctoral Academic Forum have closed. This article reviews 316 papers that have been accepted by the conference. By reviewing these papers, we reveal a new trend in the development of information science and information services, and illuminate the progress in this area to empower new quality productive forces and support the development of Chinese modernization. Those papers have covered 10 major topics, such as information work, information behavior, information services, and information users. These topics provide insights for social development from multiple fields and scenarios. The development of technologies and society expands the fields and scenarios of information science research and services, and artificial intelligence technology has become a new driving force to promote the development of traditional information science research and information work. The inheritance and development of information science and services reflect the integrity and innovation in the field. This article reveals the topics, content, and methods that have been presented in this conference, showing the characteristics and trends in information science and information services. The papers that have been accepted in the conference indicate the mission and responsibility to promote new quality productive forces and the development of Chinese modernization. 2025 Vol. 44 (3): 369-380 [Abstract] ( 4 ) HTML (1 KB)  PDF (801 KB)  ( 1 )