Match Scopus IDs for Persons Job: TroubleshootingMatch Scopus IDs for Persons Job: Troubleshooting

What

The job 'Match Scopus IDs for Persons' was added to help cleanup Scopus Author IDs in Pure. The purpose of this job is to help improve the accuracy and completeness of Scopus Author IDs by using Scopus to check author IDs in Pure and provide options to add or remove IDs if necessary. Full details on this new job are outlined in the 5.21 release notes under Match Scopus IDs for Persons: a cleanup job for author IDs.

The job can: 

• Find and add Scopus Author IDs on Persons in Pure, who do not already have their Scopus Author ID on their person record.  
• Verify that an existing Scopus Author ID in Pure is still valid. 
• Remove or replace non-valid Scopus Author IDs in Pure 

When running this job, you may encounter scenarios where content has not been touched or you may receive warnings and/or errors that are not readily understood. 

The section below reviews these scenarios and how to troubleshoot them.

Why

Perons skipped: 
The 'Match Scopus IDs for Persons job' has a list of criteria that it runs through and persons that match one (or more) of the following will be reviewed by the job:

• The job ONLY process Active (current and future) Academic Staff. All non-academic staff and and former Academics will be ignored by the job. (Detailed list is found in 5.21.0 Release notes link above).

Check if the persons skipped by the job are Active Academics by going to the Master Data menu > Person overview and add the filters 'Academic staff' and ‘Affiliation period’: current and future. 

The job then uses a combination of a person's name variants and research outputs to determine if IDs should be added and verified, or removed. It can also try match on ORCID ID if the person has an ORCID ID added in Pure. 

• ORCID ID
• Name variants and Research Outputs 
• Scopus Author ID (for accuracy check) 

If none of the above three criteria, the person will be skipped. 
Note that initials in names might also not be enough for the job to be able to find a match in Scopus. The job may be skipping the person if there are only initials and none of the other criteria to try match with persons in Scopus.  

Warnings and Errors:

WARN: Could not check Author ID in Scopus - please perform this check manually
        

• This error indicates problematic Scopus IDs and usually occurs when Scopus is called with an ID that they say has been taken over by multiple other IDs.
• Examples of problems the job encounters to trigger this warning:
  • Person has added a non-standard ID as a Scopus ID; common examples include a URL instead of an ID.
  • For a standard ID, Scopus returns multiple taken over IDs. For example, when you try to navigate to an Scopus Author ID you sometimes are re-directed to another url with a new ID which indicates that ID was taken over by another ID.
    • We are in communication with Scopus on how to best handle this, but for now we log this as a WARNING so you can handle these cases manually.

WARN: Returned an unexpected result: HTTP/1.1 500 Internal Server Error ERROR: Could not retrieve Scopus EIDs for author with ID: 22836076100 Exception details
        dk.atira.pure.api.shared.contentimport.sciverse.SciVerseServiceException: resourceKey: contentimport.error.fetch, resourceArgs: {exception=GENERAL_SYSTEM_ERROR : Error calling Solr Search Service}
        

• This error occurs when we try and look up related research output for a given Scopus Author ID. This error is more rare and upon further runs, the error should no longer occur and is likely caused by Scopus being overloaded.

More information

5.21.0 Release Notes