摘要
In order to archive and utilize the information from Chinese polar expeditions to the greatest extent, we design a novel knowledge repository, in which an automatic query model based on neural networks is proposed and a data call trigger is established to keep data consistent between polar data-sharing platforms. And in this repository, anybody can make contributions to the repository by creating or updating entries with version control and an authority control mechanism. In this paper, the data sources,data processes and network structure of this repository are described, and the keywords extraction and decision support operation are detailed. The analysis of this design＇s feasibility and applicability indicates that this knowledge repository is open, sole and authoritative for Chinese polar expeditions.
In order to archive and utilize the information from Chinese polar expeditions to the greatest extent, we design a novel knowledge repository, in which an automatic query model based on neural networks is proposed and a data call trigger is established to keep data consistent between polar data-sharing platforms. And in this repository, anybody can make contributions to the repository by creating or updating entries with version control and an authority control mechanism. In this paper, the data sources,data processes and network structure of this repository are described, and the keywords extraction and decision support operation are detailed. The analysis of this design＇s feasibility and applicability indicates that this knowledge repository is open, sole and authoritative for Chinese polar expeditions.
基金
Supported by the Basic Condition Platform of the Chinese Ministry of Science and Technology-Data Sharing Infrastructure of Earth System Science(2005DKA32300)
the Youth Innovation Fund of the State Oceanic Administration(2012621)
China Polar Science Strategy Research Fund Project(20120106)
the State Oceanic Administration Polar Science Key Lab Open Research Fund(KP201110)
Key Laboratory of Digital Ocean,SOA(KLD0201 408)
