Study On Comprehensibility Risk Check For Digital Preservation

Comprehensibility means that the content of the digital objects stored in the preservation system can be understood by the target user community of the preservation system.Only be understood by users the digital resources in the preservation system can be realized their value.However,due to the great difference between digital resources and traditional paper documents in carrier form,transmission mode and quantity,digital resources may show the characteristics of vulnerability and perishability because of the damage of carrier or the breakage of transmission link or the inability to be captured accurately and effectively in the huge digital flood,and making the comprehensibility of digital resources is vulnerable to damage in the long-term preservation process.Therefore,it is necessary to check and identify threats affecting the comprehensibility of digital objects in real-time during long-term preservation.In this paper,risk check is studied for the comprehensibility of digital objects.Firstly,it defines the meaning of comprehensibility and its three specific understanding ways,and analyzes the types of risk that may make the comprehensibility of digital object damage and affect the understanding of target user groups.Secondly,the risk metadata of comprehensibility is designed from the three aspects of digital objects,preservation events and preservation policies.And risk metadata of comprehensibility on the attributes of digital objects include digital object' internal structure information,external structure information,unitary semantic information,knowledge point semantic information,best browsing software,restore information and message digest.On the preserving events includes external structure determination,digest computation,semantic information identification,virus check,medium refresh,migration events.On the preservation policies includes internal structure information list,best browsing tool determination,virus check cycle,medium refresh frequency,migration accuracy.Thirdly,selecting CNKI as an experimental object of this study and dividing digital objects into several levels according to the group standard of category dimension,subject dimension,time dimension.Using the stratified random sampling method to extract a certain number of samples from each stratified unit and constructing the set of experimental samples.Afterwards,applying risk detection method to implement understanding risk detection experiment,using the risk check method to check comprehensibility risks,analyzing statistically the test results and the potential causes of risk,and making possible measures to reduce risks.Finally,the main work of this research are summarized,and proposed the follow-up research direction of this study.