437 Research on Mining Patent Layout Intention Based on Graph Embedding from the Perspective of Industrial Chain Hot! Zhai Dongsheng, Kan Huimin, Li Mengyang, Xu Shuo, Chen Mengmeng DOI: 10.3772/j.issn.1000-0135.2022.05.001 In the background of the knowledge economy, technological competition among countries is becoming increasingly fierce. However, there are cognitive and operational weaknesses in China's patent layout, which restrict the development of China's high-tech industry. Therefore, it is of great significance to propose a set of effective patent layout analysis methods to guide enterprises in further implementing such layouts. Currently, the analysis perspective of patent layouts is limited to the enterprise level, and the traditional layout analysis methods cannot directly show the technological attack-and-defense intention between enterprises. Based on the description of patent distribution structure of the industrial chain, this study summarizes the basic patent distribution pattern and its layout intention to realize its mining from the perspective of the industrial chain. This study first modified the organizational structure of patent knowledge by introducing the structural and functional associations at the micro level, and the domain patent knowledge map was then constructed according to the revised Patent Knowledge Map ontology. Next, we used a graph embedding algorithm to describe the patent distribution structure of the industry chain. We then summarized five basic patent distribution patterns and analyzed their layout intention, based on which we mined the layout intention of each layout subject in the industrial chain. Finally, through empirical research in the field of non-perfluorinated proton exchange membrane, we verified the effectiveness and practicability of the above method, and provided suggestions for the safe development of China's non-perfluorinated proton exchange membrane industry. 2022 Vol. 41 (5): 437-450 [Abstract] ( 314 ) HTML (1 KB)  PDF (4166 KB)  ( 405 )

451 Tracing the Knowledge Flow Main Path Based on Important Citations Hot! Xia Hongyu, Hu Qian, Wang Zhongyi DOI: 10.3772/j.issn.1000-0135.2022.05.002 The traditional main path method based on citation network analysis does not consider the knowledge contribution of the cited literature to citations. It is considered that all citations of an academic document are of equal importance to the document, even though there can be various reasons and different levels of importance from one document to another. This paper applied main path analysis, taking different important citation levels into consideration, and constructed a citation importance index as a reference variable for adjusting traversal counts. The experimental results showed that the key main path and global main path after weighted adjustment of citation importance achieved the highest Precision and F1 scores in this experiment. This research proves that the weighted adjustment with citation importance can increase the continuity of the main path link in time, improving the correlation between nodes. This improves the traceability of the main path analysis method and the ability to find key nodes. 2022 Vol. 41 (5): 451-462 [Abstract] ( 367 ) HTML (1 KB)  PDF (2432 KB)  ( 311 )

463 Recognition of Emerging Technologies Based on Dynamic Characteristics of Multi-dimensional Attributes Hot! Li Chang, Yang Zhongkai, Dong Kun DOI: 10.3772/j.issn.1000-0135.2022.05.003 Under the background of a new generation of scientific and technological revolutions, identifying the emerging technology will help understand the frontier technology and its trends and help methodically plan scientific and technological strategies. This study first delineates the definition and multi-dimensional properties of emerging technologies. Then, based on the characteristics of attribute change of emerging technologies in the temporal dimension, this study proposes an emerging technology identification method using multi-dimensional attribute change characteristics. Empirical research has been conducted in the field of carbon nanomedicine. The results of the empirical research demonstrate that the method can effectively identify emerging technology and has functions as the fine-grained display of technical content and evolution processes. 2022 Vol. 41 (5): 463-474 [Abstract] ( 356 ) HTML (1 KB)  PDF (2222 KB)  ( 309 )

475 Automatic Indexing of Large Scale Subject Words Hot! Han Hongqi, Gui Jie, Zhang Yunliang, Weng Mengjuan, Xue Shan, Yue Lindong DOI: 10.3772/j.issn.1000-0135.2022.05.004 Existing subject indexing methods can only extract words that appear in the text but cannot select the words that have strong semantic correlation and do not appear from tens of thousands or hundreds of thousands of subject words. The multi-label text classification algorithm based on machine learning needs training data under each label, limiting its application in subject indexing. Aiming at the indexing requirements of large-scale subject words in massive documents, this study proposes an automatic indexing method based on the distributed word vector technique, which uses the word vector trained by a large-scale corpus to generate representation vectors for subject words and text documents of the same dimension and realizes the calculation of semantic similarities between them. The mapping table between subject and common words is constructed based on a large-scale corpus, so that the text vector is only compared with a small number of semantically strongly related subject word vectors, which significantly reduces the amount of calculation and improves the indexing efficiency. The developed automatic indexing tool has been applied to subject indexing on nearly 100 million documents and has achieved satisfactory speed. Compared with the Jieba keywords, the proposed method has a lower coincidence degree between the subject words and author keywords and achieves better indexing accuracy than the Jieba keywords after removing the non-subject words in the Jieba keywords. 2022 Vol. 41 (5): 475-485 [Abstract] ( 387 ) HTML (1 KB)  PDF (1691 KB)  ( 401 )

486 Journal Evaluation Method Based on the Integration of Peer-Review Consensus and Bibliometric Indicators Hot! Zhang Xu, Lyu Dongqing, Ruan Xuanmin, Cheng Ying, Ju Xiufang DOI: 10.3772/j.issn.1000-0135.2022.05.005 Current research on journal evaluation lacks in-depth consideration about integrating qualitative and quantitative methods. This study takes advantage of the consensus of peer review among first-rate journals and proposes a new indicator, Journal Authoritativeness Factor (JAF), by integrating peer-review consensus and bibliometric indicators with the help of TrustRank. We also introduce the Journal Authoritativeness and Impact Index (JAII) by combining JAF and Journal Impact Factor (JIF). To verify the effectiveness of JAF and JAII, this study conducts empirical research by selecting journals recommended by the Federation of Management Societies of China (FMS) as the “gold standard” and citation data from 2000 to 2018 in the Chinese Social Sciences Citation Index (CSSCI) database as evaluation data. Results show that JAF is robust among different citation windows. Compared with the two baseline indicators, the ranking of journals by JAF and by FMS are more consistent, with the correlation between the journal ranking of JAII and FMS being as high as 0.716. 2022 Vol. 41 (5): 486-496 [Abstract] ( 299 ) HTML (1 KB)  PDF (1601 KB)  ( 222 )

497 Research on Semantic Relevance of Medical Text Oriented to Event Ontology Hot! Li Yueyan, Wang Hao, Deng Sanhong, Chen Yan DOI: 10.3772/j.issn.1000-0135.2022.05.006 With the rapid development of Internet-based medicine, digital and smart economies have become inevitable development trends of the future. The semanticization and standardization of medical knowledge is an important means to realize smart medical and digital medicine. However, the more mature medical ontology at this stage only describes some established static knowledge and cannot reveal the dynamic relationships within medical knowledge. Therefore, based on knowledge representation and organization, it is very important to construct a structured representation of medical knowledge that meets the characteristics of narrative text. This study started from the basic theory of narratology and knowledge representation of events. First, based on whether it has narrative characteristics, medical texts are divided into narrative and conceptual texts. Then, the conceptual and the narrative medical texts were semantically modeled and represented, and the medical knowledge ontology model based on the event ontology was constructed. Finally, according to the conceptual model proposed in this paper, the semantic structured representation of the SARS-CoV-2 virus invasion process is realized. The experimental results of preliminary labeling show that the migration of the event ontology model to the semantic structured description of medical text is helpful to realize the in-depth representation and knowledge discovery of medical text. This can better describe the dynamic relationships within medical knowledge and can efficiently characterize the dynamic development characteristics of medical objects in time and space. 2022 Vol. 41 (5): 497-511 [Abstract] ( 237 ) HTML (1 KB)  PDF (3345 KB)  ( 344 )

512 Influence of Subject and Region Intersection in the Knowledge Exchange of Scientific Academic Collaboration in Digital Humanities Hot! Ye Guanghui, Peng Ze, Bi Chongwu, Xia Lixin DOI: 10.3772/j.issn.1000-0135.2022.05.007 To change the practice that previous studies were limited to in exploring the influence of a single measure of discipline or region crossover on knowledge exchange of academic collaboration, this study integrates two measures of such crossovers to analyze the complex influence process of these two measures on the formation of meta-network characteristics of knowledge flow of academic collaboration. First, we construct such a meta-network, and set up an index system for the analysis of the characteristics of nodes, subgroups, and the entire network. We then analyze the characteristics of the meta-network according to the calculated index values. The two sub-networks from the two measures of geographic and subject intersection are analyzed, and the characteristics of these two sub-networks in each dimension of node, sub-group, and the whole network are obtained. Finally, the relationship between the regional and the subject cross subnets and the meta-network characteristics of academic collaboration knowledge flow is established, and the function mechanism of subject cross and regional cross in the formation and evolution of the meta-network characteristics is obtained. 2022 Vol. 41 (5): 512-524 [Abstract] ( 331 ) HTML (1 KB)  PDF (1863 KB)  ( 274 )

525 The Evolution of the Online Public Opinion of Stakeholders in Emergencies Hot! Zhang Jiaomeng, Shi Rongrong DOI: 10.3772/j.issn.1000-0135.2022.05.008 In this study, a method for monitoring public opinion on public health emergencies in conjunction with stakeholders is proposed. Taking micro-blog data on the COVID-19 epidemic as an example, 11 types of stakeholder were divided according to their social roles in the epidemic. The models of Latent Dirichlet Allocation (LDA) and LDA2vec were connected in series for topic extraction, and SnowNLP was used for text sentiment classification. Through the statistics of the absolute attention and relative attention of topics, the evolution of the online public opinion of different stakeholders was obtained. The empirical results showed that the stakeholders’ concerns were more consistent during the outbreak period, but more scattered during the stable period, then consistent as the epidemic gradually came under control. Stakeholders in the same roles showed similar topic evolution and emotional evolution, but their focuses were still different. Absolute attention reflects the concerns under the influence of mainstream public opinion, while relative attention reflects the concerns related to the stakeholders’ own interests. The study revealed the evolution of public opinion of stakeholders in public health emergencies, so as to provide a theoretical basis and decision-making references for the government to accurately monitor public opinion trends among different stakeholders in public health emergencies. 2022 Vol. 41 (5): 525-535 [Abstract] ( 418 ) HTML (1 KB)  PDF (3414 KB)  ( 632 )

536 Research on the Interdisciplinary Development of Library and Information Science Doctoral Dissertations in China and North America Hot! Song Yanhui, Zhu Li, Shu Fei, Qiu Junping DOI: 10.3772/j.issn.1000-0135.2022.05.009 As a developing subject, Library and Information Science (LIS) shows a growing tendency for being more interdisciplinary. Changes in the research focus in this field have been emphasized by many scholars. China's LIS originated in North America, but comparative research on the interdisciplinary development of Chinese and American doctoral dissertation topics is rare. This paper investigated changes in the number of LIS doctoral dissertations, the interdisciplinary situation of relevant topics, and the influence of doctoral advisors on topic selection in the period 1994-2018. The results showed that, as a highly interdisciplinary discipline, LIS doctoral dissertations involved 39 topic categories. Information Science, Library Science, and Computer Science were the most common LIS research fields in China and North America, with a close relationship. It was also found that the dominance of traditional Library Science is declining. The interdisciplinary research in Chinese LIS doctoral dissertations focused on the integration of Humanities and Social Science, while that in North America focused more on interdisciplinary research related to information technology. Another important finding was that the interdisciplinary degree of dissertation topics was deeply influenced by the academic background of doctoral advisors. The systematic research content of LIS doctoral dissertations in North America is of referential significance to LIS doctoral education in China. 2022 Vol. 41 (5): 536-548 [Abstract] ( 246 ) HTML (1 KB)  PDF (2583 KB)  ( 387 )