How to manage Name Based Search Settings in an Automatic Claiming world

Name based search settings will determine the identifiers presented to you in the automatic claiming page. You may also have a lot of Pending items that make it difficult to target which publications or grants really belong to you. If so, perhaps it's time to revisit your search settings. 

How do I refine my Name-based Search Settings?

Name Variants: Name variants should correspond to however your work is indexed in data sources. Name variants which contain two initials will retrieve more accurate results, but include a variant with one initial if you know some of your work has been indexed in that manner. Remove any name variants where you have retrieved the publications or grants already.

Address: Once you have retrieved publications from your previous institutions they no longer need to be in your search settings and can be removed.

Data Sources: If you don't work in a discipline like Medicine consider turning off PubMed. Alternately, if you are in Medicine you may not want to search MLA (Modern Language Association) or arXiv, so uncheck the boxes in datasources.

Popular Name: Consider unchecking all datasources, then clearing all your pending items. This effectively turns off Name-based Search Settings and allows you to instead use Identifiers on the Automatic claiming page. Data sources with identifiers will still be searched in the background.

If you installed Elements when Automatic Claiming was not available you may have default settings that can be reviewed.  Relaxing the complexity of the search settings and analysing your database will target where automatic claiming works best.

How do I review search settings globally?

Non-Current Users: Turn off all the data sources for these users. Create an autogroup to identify your non-current users [current]=0 and use this to bulk disable each datasource. By default Non-Current users are not automatically added to the Sync Queue, but editing an individual's search setting page overrides this default. This will remove redundant searches and reduce the number of pending items. You will still need to clear the pending items manually.

Default Search Terms for new users in Elements: If you have search terms at every Primary Group, consider enabling the Override and simplifying/reducing the address terms and datasources that are offered to new users. 

You can use bulk enable or disable datasources to further refine the data source available to groups of users. You can use autogroups very effectively in this scenario. For example, [CreatedWhen]>'2018-01-01' AND PrimaryGroupDescriptor = 'Lilliput Centre for Research Excellence'

Quick Guides: Update your documentation to focus on automatic claiming. Use ORCiD and author identifiers as the default. Encourage your users to populate search settings with name and two initials and remove address terms if the name is not popular.

Reduce the number of Pending items in your system: Large numbers of pending items create noise in your system. Use the System Admin Dashboard or the OA Monitor Reports to identify how many pending items you have. Ask for help via a Support request on how to manage large numbers of pending items.

Utilise Data Extracts from the Reporting Database:

1. Search setting report: Use the Name based Search Settings report to prioritise and sort what should be updated.

2. ORCID adoption report: Use the following report to gain an insight into how to use the Elements Reporting Database to understand ORCID adoption by your researchers. Alternatively, use the basic report 'Associated ORCID IDs'. 

1. If a user has an Author ID entered in their search settings pre-automatic claiming, following upgrade will it appear under ‘Do these identify you?’ (requiring confirmation) or under ‘External Profiles’ (already associated, requiring no confirmation)? Does ORCID behave the same as other user IDs?
The claimed identifiers will appear under "External Profiles." This applies to all identifiers, including ORCID.
2. If a users previously entered author IDs, do they default to either ‘auto-claim’ or ‘auto-suggest’?
Auto-claim.
3. Does ORCID on the automatic claiming page require authentication?
Yes and no. You may want to review the FAQ questions about ORCIDS here.
4. If an author has just one publication on which they have an associated author ID ior email will they appear in 'Do these identify you?' i.e. what is the threshold for something to be suggested?

There is no threshold. If an author only has one identifier and the search settings entered into the system match it AND if the data source has created an identifier for it (Scopus requires a certain number of publications before they'll create an identifier), then Elements will retrieve the identifier.