1129 National Security Strategy Information Demand Modeling and Intelligent Matching Research Hot! Li Gang, Yu Hui, Xia Yikun DOI: 10.3772/j.issn.1000-0135.2021.11.001 As an important part of national security information management, research on the demand for this information is key to supporting strategic decision-making and has great theoretical and practical significance for improving the overall management ability of national security strategy. This study uses concept analysis, structure analysis, and case analysis to reveal the concept connotation, attribute characteristics, and classification of national security strategic information demand. The study determines the element structure behind the national security strategy's information demand reasoning; puts forward the functional tasks involved in demand perception prediction, analysis and evaluation, and response decision-making; and builds the national security strategy information demand model and intelligent matching framework based on the information process. This provides new ideas and methods for the innovation of national security strategy information management research. 2021 Vol. 40 (11): 1129-1138 [Abstract] ( 292 ) HTML (90 KB)  PDF (1907 KB)  ( 395 )

1139 Strategic Thinking and Path Selection of Information Integration of Big Data for National Security Hot! Ba Zhichao, Liu Xuetai, Ma Yaxue, Li Gang DOI: 10.3772/j.issn.1000-0135.2021.11.002 Comprehensive information integration of big data for national security is an important measure to support national security management strategy decisions and effective responses to emergencies. It is also an inevitable requirement for responding to the national security big data strategy and the construction of “digital China.” By grasping the strategic significance and social application value of comprehensive information integration of big data for national security, we explore feasible schemes for comprehensive information integration of this data and propose specific implementation paths; this includes the theory framework of big data integration, the construction of city profile and resource pool, the planning and development of the data honeycomb system, data integration research, and demonstration for cities or regions. The comprehensive information integration of big data for national security is gradually realized from the bottom up. 2021 Vol. 40 (11): 1139-1149 [Abstract] ( 355 ) HTML (84 KB)  PDF (1572 KB)  ( 784 )

1150 Social Media Multimodal Analysis for Emergency Management Hot! Xu Yuan, Mao Jin, Li Gang DOI: 10.3772/j.issn.1000-0135.2021.11.003 Social media has become an important information source for situation awareness during emergencies. Multimodal information such as text, images, and videos viewed through social media can be widely used for emergency information management. Multimodal information analysis during emergencies is a challenging topic and has received significant attention from both academia and industry. We review studies on social media analysis in emergencies. The multi-dimensional characteristics of multimodal information in social media are analyzed using three dimensions: content, spatial-temporal, and information dissemination. The key methods and technologies of multimodal information analysis are summarized from three aspects: information acquisition, information integration, and information mining. Finally, we construct a multimodal information analysis framework for emergency management based on this review and propose future research directions on information acquisition, description, analysis, and visualization. We expect this survey to provide guidance for research on and practice of social media multimodal information analysis and to improve the ability to manage emergencies. 2021 Vol. 40 (11): 1150-1163 [Abstract] ( 397 ) HTML (156 KB)  PDF (1508 KB)  ( 1370 )

1164 Construction of National Security Event Map and Its Application for Situation Awareness Hot! Li Gang, Wang Shiyun, Mao Jin, Li Baiyang DOI: 10.3772/j.issn.1000-0135.2021.11.004 National security events related big data creates challenges for the management of those events. When a crisis occurs, determining how to integrate massive, multi-source, heterogeneous, and dynamically changing national security event-related data, and then extracting valuable intelligence so that a scenario description and situational understanding of national security events can be formed, is critical for managing national security events. Therefore, according to the intelligence needs of situation awareness during such events, this article proposes a new national security big data organization model: the National Security Event Map. We further explore its automatic construction methodology and approaches for situation awareness that can be based on the event map. The National Security Event Map can achieve knowledge representation, structured organization, and effective management of events, entities, and their relationships, which could enrich information organization theory and methodology in information science. It can also be used for comprehensive monitoring and perception of national security incidents to provide intelligence support for related management decisions. 2021 Vol. 40 (11): 1164-1175 [Abstract] ( 641 ) HTML (109 KB)  PDF (2741 KB)  ( 2045 )

1176 Financial Intelligence Studies in Digital Finance Era: Discipline Status, Discipline Connotations, and Research Directions Hot! Ding Xiaowei DOI: 10.3772/j.issn.1000-0135.2021.11.005 The actual status of Financial Intelligence Studies is very different from what it should be. This situation needs to be changed. With the advent of the Digital Finance era, the author advocates building a next-generation digital financial information foundational infrastructure, architecture, computing paradigm, and integrated innovation incubation platform based on Blockchain Driven Trustable Big Data and Trustable Artificial Intelligence (AI) Technology. The next generation of Finance characterized by Digital Finance provides rich theoretical nourishment, practical experience, and realistic scenes for the discipline of Financial Intelligence Studies and provides an excellent opportunity for original academic research. Original theoretical and practical research on Financial Intelligence Studies will feed-back and boost the vigorous development of the Next-Generation Finance. The development of the Next-Generation Finance needs the support of the original theory of Financial Intelligence Studies, which can help answer the questions such as the direction that financial innovation should take in the future. Financial Intelligence Studies can build its discipline system, academic system, and discourse system around the Next-Generation Finance. In the light of the Next-Generation Finance, the discipline connotation of Financial Intelligence Studies includes: basic cognition of Financial Intelligence discipline; handling of financial information, knowledge and intelligence; monitoring and early-warning financial risks from the perspective of Financial Intelligence; under the realistic conditions of technology empowerment, studying the Next-Generation Finance from the perspective of Financial Intelligence; and carrying out Financial Intelligence research according to the practical needs of continuous development. The systematic framework of Financial Intelligence Studies includes: foundational infrastructure level, theoretical basis level, computing paradigm level, and application practice level. The research directions of Financial Intelligence Studies in the Digital Finance era mainly include: financial risk prevention and control and financial security, regulatory technology and financial order optimization, integrity and innovation in the financial field, and the connection and game of world finance dominated by China and the United States. 2021 Vol. 40 (11): 1176-1194 [Abstract] ( 495 ) HTML (157 KB)  PDF (1246 KB)  ( 917 )

1195 Logical Link between Patent Licensing and Patent Citation and Its Application from the Perspective of Information Science Hot! Li Rui, Xiang Shuxuan, Huang Jingyun DOI: 10.3772/j.issn.1000-0135.2021.11.006 “Patent licensing” refers to the role and status of licensors and licensees in the technology market. “Patent citation” reflects the existing and potential relationships between citing and cited patents in the technical knowledge field. Thus, patent licensing and patent citation have a natural and deep-seated logical connection. Patent citation occurs before when patent rights are issued, while patent licensing may or may not occur after when patent rights are issued. It is therefore possible to predict patent licensing by analyzing the characteristic index related to patent citation. This study examines the communication industry, in which patent licensing is a common practice. We measured 4,985 licensing patents by 540 licensors and their 1,332 forward citations together with 6,876 backward citations. Next, we analyzed the correlations between the variables related to patent citation and the probability of patent licensing. We then attributed the correlations of these variables to the principles of innovation economics and constructed the patent license prediction index using the Brovey converter model, which was proved to be effective through experiments. 2021 Vol. 40 (11): 1195-1208 [Abstract] ( 378 ) HTML (161 KB)  PDF (3135 KB)  ( 447 )

1209 Detection of Scientific Knowledge Structure Based on Graph Representation Learning Hot! Liu Feifan, Zhang Shuang, Luo Shuangling, Xia Haoxiang DOI: 10.3772/j.issn.1000-0135.2021.11.007 Accurately identifying and detecting scientific knowledge structures is of fundamental importance for understanding the development of subdisciplines in a certain field, formulating science and technology policies, and conducting research management activities. The current methods implemented by researchers to address this challenge mainly focus on two aspects: text mining and social or complex network analysis. However, few studies have fully integrated the information obtained from these two methods, and they are often only used as a basis for cross validation. Therefore, in this study, we use the advantages of graph deep learning methods emerging in the field of deep learning and propose a research framework for scientific knowledge structure detection based on deep graph neural network models combined with document representation and manifold learning algorithms. Two datasets were selected to validate the proposed research framework that are representative of basic research disciplines and new emerging research fields, respectively. The experimental results show that graph deep learning can effectively integrate the topical feature information of the literature and citation relationship feature information, thereby detecting a clearer domain knowledge structure. This study expands the application scenarios of the graph neural network model and presents reference value for the application of scientific and technical information engineering. 2021 Vol. 40 (11): 1209-1220 [Abstract] ( 335 ) HTML (108 KB)  PDF (6236 KB)  ( 630 )

1221 On the Concepts and Approaches of Computable Knowledge in Biomedical and Health Sciences Hot! Du Jian, Kong Guilan, Li Pengfei, Bai Yongmei, Zhang Luxia DOI: 10.3772/j.issn.1000-0135.2021.11.008 “Computable knowledge” focuses on transforming human-readable knowledge into machine-executable forms by extracting and programming processes on digital knowledge objects. It can be regarded as the “keystone” in supporting the massive knowledge application in the cycle of learning health systems, i.e., “from data to knowledge, from knowledge to practice, and then from practice to data.” This concept has become a new field of research in health data science, and it also provides a new paradigm for digital library and knowledge computation research in the field of library and information science. This study proposes two approaches to making medical knowledge computable. One is a data-mining-driven approach. Computable knowledge can be extracted from the data in tables of medical literature, expressed in code, and managed in the Knowledge Grid (K-Grid). For example, a machine-executable version of the predictive model can be encoded in any appropriate computer language. When given an instance of data about an individual, this encoded model can quickly and accurately generate a risk prediction or useful advice. The second is a text-mining-driven approach that extracts Subject-Predicate-Object (SPO) triples from an unstructured text, such as the assertions in clinical guidelines and medical literature. By incorporating the evidence and data into a given SPO triple, we can calculate the confidence score for such a knowledge unit. The SPO triples can be stored in graph databases (K-Graph) for automatic question answering for a specific condition, such as treatment recommendations ranked by the confidence level to support medical intervention decision-making. Several challenges for the development and application of computable medical knowledge have been discussed. We hope to introduce an interdisciplinary approach to investigating computable medical knowledge and provide conceptual and technological preparations for the learning health system in China. 2021 Vol. 40 (11): 1221-1233 [Abstract] ( 292 ) HTML (114 KB)  PDF (2098 KB)  ( 537 )

1234 Chinese Disease Name Normalization Based on Multi-task Learning and Polymorphic Semantic Features Hot! Han Pu, Zhang Zhanpeng, Zhang Wei DOI: 10.3772/j.issn.1000-0135.2021.11.009 In order to solve the problem of a large number of disease designations in online texts, a Chinese disease name normalization model based on multi-task learning and polymorphic semantic features (multi-task attention-dictionary BERT GRU-CNN, MTAD-BERT-GCNN) is proposed. First, word2vec and Glove were used to generate external semantic feature vectors that integrate local and global semantics. Second, CNN and BERT were used as benchmark models for comparative experimental analysis. Third, GRU, LSTM, BiGRU, and BiLSTM were introduced on CNN to extract semantic relationships between texts. Next, from the perspective of multi-task learning, the above model was combined with BERT to capture static and dynamic semantic information. Finally, the medical dictionary was introduced to calculate the attention matrix as an auxiliary task to adjust the static vector, thereby further improving the model effect. Our experiments were carried out using the self-built Chinese disease name normalization dataset, ChDND. The experimental results found that the MTAD-BERT-GCNN model achieved 89.60% accuracy on the Accuracy@10, which is higher than the basic word-level CNN, and the word-level CNN increased by 12.96% and 5.12%, respectively. This research introduces the concept of multi-task learning in the normalization task of Chinese disease names and optimizes it from the level of the semantic vector and model framework, which has good application value in the construction of Chinese medical knowledge graphs, information extraction, and natural language understanding. 2021 Vol. 40 (11): 1234-1244 [Abstract] ( 218 ) HTML (137 KB)  PDF (2417 KB)  ( 727 )