Subjects: Library Science，Information Science >> Library Science submitted time 2023-08-26 Cooperative journals: 《图书情报工作》
Abstract： [Purpose/significance] With the substantial growth of the number of academic articles, researchers usually spend more time in searching, acquiring and understanding the content of academic articles under the current publishing modes. In order to facilitate scientific information can be disseminated and exchanged rapidly, a new publishing mode:semantic publishing, which focuses more on the fine-grained content of academic journals, is arising in academic communities. This paper intends to introduce a representative semantic publishing mode:"nanopublication", and explore the possibility and characteristics of application in different subject fields. [Method/process] Firstly, we introduced the nanopublication model.Secondly, we reviewed the current status of nanopublication's application by literature review. Finally, we analyzed the characteristics of nanopublication's application in different subject areas with examples. [Result/conclusion] The research shows that:①Nanopublication is mainly used in biomedical field currently, seldom used in the fields of computer science and human science, and hardly used in other fields; ②It is possible to extend nanopublication to other subject fields, and construct nanopublications according to the characteristics of different subject fields.
Subjects: Library Science，Information Science >> Library Science submitted time 2023-08-26 Cooperative journals: 《图书情报工作》
Abstract： [Purpose/significance] Semantic annotation of digital images is an effective way to solve this problem. The foundation of semantic image annotation is the construction of semantic models. This paper intends to review the existing mainstream semantic models for image annotation, and explore their advantages and disadvantages.[Method/process] Firstly, four representative semantic models for image annotation were reviewed, including Eakins model, Jaimes & Chang model, Kong model and Panofsky model, using literature survey, and then the first three models from three aspects (i.e. semantic level, extensibility and application range) were compared and analyzed using comparative analysis.[Result/conclusion] Through the above analysis, it can be concluded that Eakins model has the most comprehensive semantic level, the strongest semantic expression ability and the widest application range, whereas Kong model is the most scalable and adaptable one.
Subjects: Library Science，Information Science >> Information Science submitted time 2023-04-01 Cooperative journals: 《图书情报工作》
Abstract： [Purpose/significance] Relation prediction is an important task in knowledge graph completion, and plays an important role in improving the completeness of knowledge in knowledge graphs. The paper proposes a new relation prediction method that combines internal structure features and external text features, which aims to predict the missing relations between two entities in knowledge graphs. [Method/process] The method transforms the relation paths in a knowledge graph and the texts that involve entity relationships into matrixes, learns the structure features and text pattern features related to a specific relation type through convolutional neural networks, and then trains a model based on the learned features for relation prediction. [Result/conclusion] The results shows that the performance of our proposed method on evaluation data sets is superior to the state-of-the-art approaches, and the method can effectively improves the performance of knowledge graph relationship prediction. Through practical application, it wvas found that this method has high application value in knowledge services.
Subjects: Library Science，Information Science >> Information Science submitted time 2023-04-01 Cooperative journals: 《图书情报工作》
Abstract： [Purpose/significance] Argument mining can identify the argument structure in argumentative texts, so as to help users to understand the reason and process of drawing a conclusion, and thus has important academic and application value. In recent years, argument mining has obtained great attention in social media content mining, legal assistance judgment, decision support and so on, and become a new research direction in the field of text mining. The purpose of this paper is to sort out and summarize the existing studies and application of argument mining, to discover new research hot spots, and to provide reference for future research. [Method/process] We serched literatures by using the keywords of "argument mining OR argument component OR argument structure OR argumentation mining" from the Web of Science and ACL databases and obtained a total of 220 articles, and then analyzed them from three aspects:argument models, argument mining tasks and argument mining applications by intensive reading and content analysis. [Result/conclusion] The research on argument mining has just started. Existing studies focused more on simple argumentative texts such as social media, and ignored complex argumentative texts such as scientific papers. In future, researchers can focus on the argument mining of complex texts and carry out research from three aspects:argument annotation schemas, the identification of argument components and relationships, and the optimization of argument structures.
Hits
Hits
Downloads
Comment