Automatic claiming in Elements

Elements uses a range of researcher identifiers -- including the arXiv Author Identifiers, Dimensions Researcher IDs, figshare.com accounts, ORCIDs, Web of Science ResearcherIDs, Scopus IDs, SSRN Author IDs and email addresses -- to facilitate the claiming of publication and grant data.

Identifier-based searching improves document retrieval and greatly reduces the amount of time it takes for users and curators to process pending publications and grants.

To begin, navigate to Profile & Work > Personal Profile & Identifiers > Automatic Claiming (in 5.x: Menu > My Account > Automatic claiming).

On the Automatic claiming page, there may be a list of data sources under the "Do these identify you?" section. Elements has harvested these identifiers from data sources based on search settings already in the system and an algorithm that looks for identifiers associated with harvested publications. 

Click on the Yes, No or Ignore button beside each identifier to tell Elements what it should do when it finds this identifier in a publication. If unsure of a particular identifier, click on the identifier's link to check the identifier's profile at its data source.  

Note: The 'Auto reject' and 'Ignore' options can be accessed by clicking either the 'No' or 'Ignore' button. The difference between the two buttons is which option is offered as the default.

Helpfully, some data sources list other identifiers, which can help confirm which identifiers are associated with a researcher. For example, Daniel Hook's ORCID page also lists his Scopus ID and, likewise, his Scopus page shows his ORCID.

Clicking the Yes button for an identifier will present a confirmation screen that explains exactly what will be done with the identifier if accepted.

The default option is to "Auto claim items" which means publications or grants with the author identifier will be automatically added to the "Mine" list. The other option is to ask Elements to "Auto suggest items," which places the publications into the "Pending" list.

If claimed, the newly verified identifier will be added to the list of External profiles farther down the page.

If a known identifier is not listed in the suggestions, go the Add external profiles section and click on the database name.

This will open a dialogue box where the identifier can be entered.

If the identifier is recognised, Elements will retrieve a sample set of articles from the database. Click the Verify button and Elements will retrieve all of the items associated with the identifier.

To manage or remove an identifier, click the > symbol on the right side of the page.

Then, click the Manage button to modify how the identifier is handled.

Email addresses can also be used to autoclaim publications. The email address associated with an Elements account will be automatically added to the autoclaiming system, but additional addresses can be entered. First, click the hypertext link under Add email addresses and then enter an email address. 

Elements will also review the metadata in claimed publications and suggest other email addresses if others are found. As with publication identifiers, publications can be pushed into the "Mine" or "Pending" folders.

• Settings for ORCID and figshare have been moved to the Automatic claiming page.

• Authors have the option to search a data source using only their researcher identifiers by unticking the Search by name checkbox. This will particularly useful for researchers with very common names who are reasonably sure most or all of their publications are associated with their researcher identifiers.

As an example, the arXiv database does not index author affiliations (address information), but it does have a robust identifier system. Authors who typically see a lot of false positive hits from arXiv in their Pending list are encouraged to enter their Author identifier and untick Search by name.

How does the synchronizer harvest identifiers? My profile has retrieved several identifiers where the last name is correct, but the first name only shares the first initial.

The synchronizer examines name-based search settings (Name Variants & Addresses) and retrieves probable matches based on them. If the author's name-based search settings are relatively simple (Last Name, Single Initial) and if the author's surname is fairly common, a lot of identifiers could be retrieved. Recommendations to mitigate include: 1.) Using two initials when practical (not all authors publications are indexed with two initials) and also the full first name. 2.) Enter Address terms. These will also restrict the number of identifiers retrieved. After saving the new search settings, clear the Pending queue and rerun the search.

I have more than one Scopus Author identifier. Can I enter all of them?

Yes, although the addition of multiple identifiers of some other types is disallowed.

What happens if one of my Author identifiers is merged or deleted by the provider?

Nothing deleterious will happen to the publications data in Elements if an identifier is deleted by the provider. If a new identifier is created, eventually Elements will retrieve it and present it for confirmation.

Will I still have to confirm my ORCID is mine by entering my password?

Autoclaiming has meant that Elements integration with ORCID has changed significantly.

• Claiming an ORCID, but not connecting the account by entering a password, will retrieve documents from data sources that have the ORCID in their metadata, including Dimensions, Scopus, Europe PubMed Central and Web of Science.

• Connecting an Elements account to an ORCID account via the ID/PW interop will allow Elements to retrieve additional data from ORCID records that can then be used to search other data sources, though please note Elements does not download full publication records directly from ORCID due to highly variable data quality.

• If an author has not authenticated their ORCID account, the Account not connected will appear in the External profiles section and Search errors will be logged in the System Administrator's dashboard.

Can Author identifiers be harvested from the Reporting Database?

Yes. Identifiers are logged in the reporting database in the [User Identifier Association] table. This enables customers to see all identifiers associated with an author, including multiple Scopus IDs.

Can the "Bulk enable/disable search settings" functionality be used on ORCID and figshare?

No, the Bulk enable/disable functionality will only work on name-based data sources. To enable or disable ORCID or figshare in bulk, the System Administrator can remove access to ORCID or figshare for any Primary Group via the Publication module settings.

I claimed a number of publications then turned off my name based search settings. Will unclaimed items with associations to the author IDs remain in my Pending tab and can they be cleared?

Until an author take action by selecting 'Yes,' 'No,' or 'Ignore,' the unclaimed items will remain in the Pending folder and cannot be cleared.

An incorrect author identifier was inadvertently claimed, but when the identifier was changed to "Not mine," the publications did not clear out of the Claimed folder. What can be done?

With improvements in the automatic claiming functionality, claiming publications that have an author name that does not match the search settings should be very difficult, though not impossible. Even after setting the identifier to "Not mine," the publications must be manually rejected.

Must an author have a particular number of publications or grants in order for an author identifier to be suggested?

If no other identifiers have been claimed, an identifier will need to be found in three objects (publications or grants) in order to be suggested. 

Is every author identifier suggested in the [User Identifier] table in the Reporting Database presented to authors for claiming in the Elements UI?

No. An identifier may fall below the threshold to be presented in the UI, but could still be found in the [User Identifier] table.

I claimed an identifier but found there were still a few pending publications associated with it? Why is this happening?

In order to avoid erroneously retrieving publications, the synchronizer is programmed to not claim publications where it finds author names or address terms that are significantly different from those entered into the Name-based Search Settings. For example, if an author's surname is entered in the Search Settings as 'Palazoglu' while the publication lists the name as 'Palazoǧlu' (with the breve over the g), the latter publications would not be auto-claimed.