Intelligence Theories and Methods

777 Collaborative Dual-Chain Data Organization Model for Industrial, Scientific, and Technological Innovation from a Science-Technology-Industry Integration Perspective Hot! Zheng Rong, Wu Qinke, Hu Jing, Wei Zhongbao DOI: 10.3772/j.issn.1000-0135.2026.06.001 This study investigates a dual-chain data collaboration model for industrial, scientific, and technological innovation under the convergence of industrial and technological advancements. Focusing on a multi-source data environment, this paper adopts a literature analysis approach and follows the analytical framework of “concept clarification-logical analysis-model construction-case analysis” to review relevant literature and clarify the connotation of dual-chain data. Combining the data value chain theory with the “science-technology-industry” framework, this study elaborates the collaborative organization logic and constructs a model based on this logic, which is further analyzed from the perspectives of collaborative elements and organizational processes. The case analysis method is used, considering the new energy vehicle industry as an example, to provide a data collaborative organization framework for the industry. The results indicate that the collaborative organization of dual-chain data follows a logical framework of “merging and integration, collaboration and linkage, and knowledge advancement.” This framework encompasses the collaborative elements—subject, object, technology, and environment—supplemented by the integration and application of databases, blockchain, data lakes, and knowledge graphs to build a collaborative organizational structure. This study proposes a dual-chain data collaborative organization model for industrial, scientific, and technological innovation and employs the new energy vehicle industry as a case study to validate its effectiveness and feasibility. 2026 Vol. 45 (6): 777-792 [Abstract] ( 19 ) HTML (150 KB)  PDF (4472 KB)  ( 18 )

793 Construction and Empirical Investigation of Early Identification Methods for Disruptive Technologies Based on Weak Signal Perception Hot! Wang Zhengyuan, Jin Junbao, Cao Kun, Zheng Yurong, Bai Guangzu, Cai Xin DOI: 10.3772/j.issn.1000-0135.2026.06.002 As the new era of technological revolution and industrial transformation gathers momentum, the early identification of disruptive technologies has become critical for nations to seize strategic high ground and reshape industrial competition landscapes. This study focused on the key segments of the scientific innovation chain (frontier research-basic research-applied research). By constructing thematic citation networks across these segments based on inter-document citation relationships, it tracked changes in triplet structures within the network over consecutive periods to identify weak signals of potential disruptive technologies. It then established a monitoring indicator system for early-stage disruptive technology signals, calculated the disruptive potential index for seed technologies, and incorporated large language models for semantic enhancement and noise filtering. This approach enabled precise identification and screening of early-stage disruptive signals within specific domains. This framework was applied to the hydrogen fuel cell field, identifying five major disruptive research directions including electrolyte and membrane technologies. The method was validated through a retrospective literature analysis, enabling the dynamic monitoring of disruptive weak signals across the entire chain from fundamental to applied research. The results demonstrate that the proposed framework significantly enhances weak signal resolution capabilities, identifying more disruptive technology directions with potential societal impact. 2026 Vol. 45 (6): 793-808 [Abstract] ( 15 ) HTML (250 KB)  PDF (4613 KB)  ( 10 )

809 Identification of Key Core Technology Innovation Opportunities by Integrating LangChain and Multi-dimensional Technology Innovation Maps Hot! Liu Peng, Wei Chenyu, Zhang Ke, Zhou Wei DOI: 10.3772/j.issn.1000-0135.2026.06.003 A comprehensive innovation breakthrough in key core technologies is a crucial foundation for building a manufacturing powerhouse and innovative nation, as well as a core driver for promoting high-quality economic development. To accelerate the pace of innovation in key core technologies, this study employed exponential random graph models (ERGM) to explore the nodal relationships among technological innovation elements in patents and conduct prompt instruction training for innovation scheme generation using large language models, thereby providing a more efficient and intelligent approach for identifying innovation opportunities in key core technologies. First, key core technologies were identified based on patent citation networks, and an evaluation system was constructed for key core technology indicators. Second, the BERTopic-ERGM model was used to mine potential combinations of innovation elements from the selected patent texts. Third, by integrating the LangChain framework with a multidimensional technological innovation map, domain-specific prompt instructions for innovative schemes were designed to enable the automatic identification of innovation opportunities in key core technologies. Furthermore, poorly generated technical solutions were analyzed in depth to further improve the overall model performance. Finally, the effectiveness of the model was validated using a dual evaluation method. The results showed that the proposed BERTopic-ERGM and LangChain-multidimensional technological innovation map model can efficiently identify innovation opportunities in the field. Compared with traditional technology forecasting methods, the proposed method not only captures emerging technological opportunities and trends more accurately, but it also enables a deeper analysis of complex interrelationships and evolutionary pathways among technologies. 2026 Vol. 45 (6): 809-827 [Abstract] ( 17 ) HTML (217 KB)  PDF (3425 KB)  ( 13 )

828 Evaluation Framework and Index System for Technological Capability Gaps Oriented to Sci-Tech Security Hot! Yuan Lin, Zhao Xiaoyuan, Zhang Guilan DOI: 10.3772/j.issn.1000-0135.2026.06.004 Against the backdrop of intensifying global sci-tech competition, accurately identifying the shortcomings in the technological capabilities of a country has become crucial for safeguarding sci-tech security. Existing evaluation models typically focus on macro-level factors and cannot fully cover the entire process of converting technology into productive forces. A theoretical analysis framework was constructed based on the innovation chain and disruptive innovation theory to explore the differences in technological capabilities between China and other countries in key fields. This framework included four links—basic research, applied research, trial maturation, and market diffusion—and two dimensions—original innovation and sustained competition. On this basis, the grounded theory was used to refine the risk early warning logic for gap evaluation and cause analysis, and a core index system was systematically selected from multi-source data. Using artificial intelligence as an example for empirical verification, the system accurately identified the structural characteristics of China’s AI sector, namely, strong sustained competitiveness yet weak original innovation. The technological capability gap evaluation system oriented to sci-tech security has advantages such as result orientation and risk early warning. It can map sci-tech security risks in key fields to specific links in technological development and dimensional shortcomings, providing decision-making references for the country to optimize the layout of sci-tech strategies and prevent security risks. 2026 Vol. 45 (6): 828-839 [Abstract] ( 12 ) HTML (121 KB)  PDF (1462 KB)  ( 6 )

840 A Complex Data Citation Index to Trace Data Reusability Hot! Zhang Lili, Hui Jiayi, Liu Ruilin, Hu Yike DOI: 10.3772/j.issn.1000-0135.2026.06.005 Tracing data reusability is crucial for open data. Recognizing the importance of data citation, this study proposes a comprehensive framework for measuring data reusability, known as the “C3 Framework”. Based on the intensity of data citations and their temporal and spatial characteristics, the framework is quantified as the Complex Data Citation Index, comprising sub-indicators such as the data citation intensity index, the data citation popularity index, and the data citation breadth index. We proposed a validation test based on the 30 most-cited data papers published in Earth System Science Data (ESSD), a journal in Earth Sciences, along with their 15812 citing papers and a corpus of 25585 data citation records. The statistics demonstrated that this method effectively interpreted discrepancies in data citations, particularly regarding citation intensity, popularity, and breadth among the samples. Further reliability tests indicated that the sample dataset passed the multicollinearity test, signifying the independence of the three citation sub-variables. Moreover, the C index correlated strongly with the current citation scales, suggesting that this new method is consistent with real-world observations. Furthermore, this method is adaptable to various scenarios and is sufficiently robust for datasets with up to 30% missing values. In summary, this study presented a user-friendly framework for interpreting data reusability based on data citations. It offers a relatively complete view of data reusability by tracing citation chains, providing an effective metric for monitoring the performance of open data work, and bridging gaps while highlighting potential opportunities. We hope the C3 solution will also benefit open data efforts. 2026 Vol. 45 (6): 840-854 [Abstract] ( 19 ) HTML (238 KB)  PDF (3731 KB)  ( 7 )

855 Research on the Participatory Construction of Trusted Data Spaces from the Perspective of Resource Orchestration Theory Hot! Chen Xiaoyu, Liu Junjie, Pei Lei DOI: 10.3772/j.issn.1000-0135.2026.06.006 Amid the accelerating process of data factorization and the strategic imperative of trustworthy data governance, trusted data spaces have emerged as key infrastructures to promote compliant data sharing and unlock the value of data elements. However, current practices for the construction of trusted data spaces are hindered by weak multi-subject collaboration, inefficient resource coordination, and unclear organizational mechanisms. While existing studies mainly focus on technical architecture or institutional frameworks, they lack a systematic perspective on participatory construction from a resource orchestration standpoint. This study introduced the Resource Orchestration Theory (ROT) as an analytical framework, and examined the current status and governance challenges of trusted data spaces in China. It elaborated on the three core processes of the theory—resource structuring, bundling, and leveraging—and argued its suitability for coordinating heterogeneous data resources across multiple actors. Building on this foundation, this study constructed a resource orchestration logic chain for trusted data spaces and proposed a participatory construction pathway encompassing three stages: resource structuring (data sovereignty confirmation and credibility evaluation), resource bundling (multi-subject collaboration to enhance data value), and resource leveraging (scenario-driven dynamic invocation). It further delineated the dynamic coordination among progressive construction stages, multi-dimensional spatial levels, and collaborative construction actors. A three-phase model—preparation, adjustment, and continuous development—was designed to support full-process governance and adaptive implementation. To empirically validate the proposed theoretical pathway and stage model, this study selected representative domestic and international cases—including the National Basic Science Data Center, Zhangjiakou Trusted Data Space, and Catena-X Automotive Network—for a comparative analysis. It systematically examined the specific manifestations and practical outcomes of participatory construction mechanisms across different evolutionary stages. The findings suggest that the ROT provides a coherent theoretical lens and operational framework for the participatory construction of trusted data spaces, enhancing resource utilization efficiency, adaptive resilience, and collaborative value co-creation. 2026 Vol. 45 (6): 855-867 [Abstract] ( 16 ) HTML (111 KB)  PDF (2222 KB)  ( 16 )

Intelligence Technology and Application

868 Impact of Organizational Knowledge Characteristics on the Formation of Technological Collaboration Relationships: A Multilevel Network Analysis in the Field of Artificial Intelligence Hot! Lin Ping, Tao Chengxu, Wu Jiang DOI: 10.3772/j.issn.1000-0135.2026.06.007 Knowledge is the foundation for an organization’s competitive advantage, and technological collaboration between organizations is a process of knowledge acquisition and integration. Exploring the impact of organizational knowledge characteristics on the formation of technological collaborative relationships can help organizations accurately identify potential partners, deepen collaboration, and enhance collaborative innovation performance. This study used patent data in the field of artificial intelligence from 2013 to 2021 to construct a knowledge-organization multilevel network. A multilevel exponential random graph model was employed to investigate the influence of organizational knowledge characteristics on the formation of inter-organizational technological collaboration relationships. The study determined that the technological collaboration network exhibited significant structural transitivity, with organizations tending to establish partnerships with their partners and leveraging trust transmission mechanisms based on shared ties to mitigate potential collaboration risks. The strength of knowledge ties, knowledge diversity, and knowledge proximity significantly drive the formation of technological collaboration relationships. However, knowledge combinatorial opportunities have a significant negative impact on inter-organizational technological collaboration. Furthermore, inter-organizational technological collaboration exhibits an evident preference for domain homogeneity, with both parties seeking partners within the same technological field. This study reveals the influence of organizational knowledge characteristics on the formation of technological collaboration relationships from a multilevel network perspective, validates the applicability of multilevel network analysis in the field of library and information science, and provides new insights and empirical support for understanding the evolution of complex technological networks. 2026 Vol. 45 (6): 868-880 [Abstract] ( 9 ) HTML (202 KB)  PDF (1779 KB)  ( 10 )

881 Fine-Grained Clustering of Scientific Fund Topics Based on Multi-task Heterogeneous Graph Representation Learning Hot! Guan Zhengyi, Xie Jing, Liu Jianhua DOI: 10.3772/j.issn.1000-0135.2026.06.008 The fine-grained topic clustering of scientific funding projects is of considerable significance for analyzing scientific funding layouts and optimizing resource allocation. To address the challenges of sparse features, high terminological density, and the lack of contextual information in funding texts, this paper proposes a fine-grained topic clustering method for scientific funds based on multi-task heterogeneous graph representation learning. The method first integrated a keyword co-occurrence network, literature citation topology network, and disciplinary classification hierarchy to construct a three-dimensional heterogeneous semantic graph. Subsequently, by combining the local feature extraction capability of graph convolutions with the global semantic association advantage of multi-head attention, a multi-task collaborative training mechanism was designed to learn the deep semantic representations of texts. Considering funded projects in the millimeter-wave domain as an example, the experimental results demonstrated that the proposed method significantly outperforms baseline models across multiple evaluation metrics. The constructed semantic representation space exhibits high intraclass compactness and interclass separation, effectively enabling fine-grained topic identification. Furthermore, the clustering results are highly consistent with the policy orientation of the fund guidelines. This method demonstrates a strong application value in the multi-dimensional text analysis of scientific funds, providing an effective tool for fund text mining and fine-grained project analysis. 2026 Vol. 45 (6): 881-895 [Abstract] ( 14 ) HTML (173 KB)  PDF (3046 KB)  ( 6 )

896 Patent Transaction Recommendation via Integrating Large Language Models and Graph Representation Learning Hot! Zhang Zhihao, Huo Yilin, Yuan Jiaxu, He Xijun DOI: 10.3772/j.issn.1000-0135.2026.06.009 Patent transaction recommendation is a crucial pathway for promoting the transformation of scientific and technological achievements. However, it faces dual challenges: highly sparse transaction data and the difficulty in effectively modeling the complex interaction relationships between organizations and patents. To address these issues, this study proposes LLM-GRL-PTR, a novel patent transaction recommendation (PTR) model that integrates large language models (LLMs) and graph representation learning (GRL). First, to mitigate data sparsity, we proposed an LLM-based data augmentation method that leveraged LLMs to infer organizational demand preferences from sparse historical transaction records and screened potential matching patents, thereby enabling personalized data completion for long-tail organizations. Second, we constructed a GRL-based recommendation framework that jointly modeled three types of relationships: enhanced organization-patent interactions, multidimensional proximities among organizations, and upstream-downstream technical linkages among patents—reflecting the industrial chain synergy. Accordingly, we built three graphs depicting heterogeneous organization-patent interaction, interorganizational proximity, and patent upstream-downstream relationship. By integrating graph neural networks with attention mechanisms, the model fused explicit interaction signals and implicit relational features to learn low-dimensional representations of organizations and patents to ultimately predict potential transaction likelihoods. Finally, empirical studies were conducted in two critical domains—electronic information and chip manufacturing. The results showed that the proposed model significantly outperformed baseline methods in terms of recommendation accuracy, and the recommended patents exhibited broad technical coverage, effectively balancing accuracy and diversity. This study provides an effective tool for intelligent patent transaction recommendation and demonstrates a promising paradigm for integrating large models into technology intelligence services. 2026 Vol. 45 (6): 896-910 [Abstract] ( 22 ) HTML (233 KB)  PDF (3284 KB)  ( 6 )

Intelligence Reviews and Comments

911 Short-Video Dissemination Risks and Governance of Emergency Incidents under Self-Media Network Contexts: Overview and Prospects Hot! Han Chuanfeng, Li Jingqi, Zhao Rongyong, Meng Lingpeng, Zhu Wenjie DOI: 10.3772/j.issn.1000-0135.2026.06.010 In a highly digitalized and intelligent network society, the propagation of emergency events is evolving from traditional newsroom reporting models to mass-oriented, short-video user-generated content models. As a core medium in the information dissemination revolution, short videos have profoundly reshaped the risk generation and diffusion mechanisms of emergencies. This study innovatively constructed an integrated “risk evolution - governance response” framework from the dual process-element perspective of general systems theory, unifying dynamic risk evolution characterization with static governance structure deconstruction. The process perspective revealed that risks follow a progressive evolutionary pattern of content alienation, system catalysis, and a social loss of control, thus synthesizing studies on risk evolution, dissemination mechanisms, and risk assessment. Further, the tripartite conflict framework of the government, platform, and user was analyzed from the element perspective, integrating institutional, technological, cultural, and collaborative governance strategies. Finally, this paper outlines research directions, including online-offline integration research, the dynamic gaming of governance subjects, and generative AI-driven governance transformation. 2026 Vol. 45 (6): 911-924 [Abstract] ( 17 ) HTML (197 KB)  PDF (2495 KB)  ( 10 )